modes/track/

[giscus[bot]]

modes/track/

Learn how to use Ultralytics YOLO for object tracking in video streams. Guides to use different trackers and customise tracker configurations.

https://docs.ultralytics.com/modes/track/

[AdadAlShabab]

Can I run two models simultaneously in one video?
Ex:
model = YOLO('E:/Ultralytics/best.pt')
model = YOLO('yolov8m-pose.pt')
results = model.track(source="file.mp4", show=True, conf=0.3, save=False)
results = model.track(source="file.mp4", show=True, conf=0.3, save=False)

I want that two models will works simultaneously with cumulative results? Is it possible? Please let me know. Thanks in advance!!

[pderrenger]

@AdadAlShabab yes, you can run two models simultaneously on one video, but you'll need to manage the inference in a way that allows both models to process the video frames concurrently. Here's a simplified example using threading to achieve this:

import threading
from ultralytics import YOLO

# Load the models
model1 = YOLO('E:/Ultralytics/best.pt')
model2 = YOLO('yolov8m-pose.pt')

# Function to run tracking
def track_video(model, source, conf, show, save):
    results = model.track(source=source, show=show, conf=conf, save=save)
    # Process results here or return them

# Create threads for each model
thread1 = threading.Thread(target=track_video, args=(model1, "file.mp4", 0.3, True, False))
thread2 = threading.Thread(target=track_video, args=(model2, "file.mp4", 0.3, True, False))

# Start the threads
thread1.start()
thread2.start()

# Wait for both threads to finish
thread1.join()
thread2.join()

This script will run both models in separate threads. If you want to combine the results from both models, you'll need to implement a mechanism to synchronize and merge the results from both threads. Keep in mind that displaying the results (show=True) from two threads might not work as expected, and you might need to handle the visualization separately.

Remember to handle the GPU memory usage carefully, as running two models simultaneously will require more memory. If you encounter any issues, consider running the models sequentially or on separate GPUs if available.

[Rid1Whitehead]

Hi,

First of all, I have been loving working with Yolov8. Great tool!

However, I have been having difficulties with a certain task. I want to use model.track on videos that I have, and then use save_crop = True, but save with a naming convention where I can track each persons ID. Currently, save_crop just gives me the cropped images of the objectes detected, but there is not way to know from which frame of the video are the crops, also, which ID is attached which cropped image. The visualization through cv2.imshow shows the IDs accross the different frames, but I cant find a way to save them. The naming convention I am looking for is something like this: "frame_30_ID_1.jpg" My current code looks something like this:

from ultralytics import YOLO
import cv2

model = YOLO("yolov8n.pt") # load model

video_path = "path/to/video.mp4"
cap = cv2.VideoCapture(video_path)

ret = True
frame_count = 0 # Initialize a frame counter

while ret:
ret, frame = cap.read()
frame_count += 1 # Increment the frame counter

# Check if the frame is the 30th frame
if ret and frame_count % 30 == 0:
    results = model.track(frame, persist=True, save_crop = True, save_txt = True)
    frame_ = results[0].plot()

    cv2.imshow("frame", frame_)
    if cv2.waitKey(25) & 0xFF == ord('q'):
        break

cap.release()
cv2.destroyAllWindows()

Any help would be greatly apprecitated! Thanks!

[pderrenger]

@Rid1Whitehead hello! I'm glad to hear you're enjoying working with YOLOv8! 😊

For your task of saving cropped images with a specific naming convention that includes the frame number and ID, you'll need to customize the saving part of the code. The save_crop functionality in YOLOv8 does not currently support custom naming directly related to frame and ID out of the box.

Here's a modified version of your code that manually saves the cropped images with the desired naming convention:

from ultralytics import YOLO
import cv2
import os

model = YOLO("yolov8n.pt")  # load model

video_path = "path/to/video.mp4"
cap = cv2.VideoCapture(video_path)

ret = True
frame_count = 0  # Initialize a frame counter
save_dir = "path/to/save/crops"  # Directory to save the crops
os.makedirs(save_dir, exist_ok=True)  # Create the directory if it doesn't exist

while ret:
    ret, frame = cap.read()
    frame_count += 1  # Increment the frame counter

    if ret:
        results = model.track(frame, persist=True)
        for i, det in enumerate(results[0].boxes.xyxy):
            # Extract bounding box coordinates
            x1, y1, x2, y2 = map(int, det[:4])
            crop = frame[y1:y2, x1:x2]  # Crop the detected object

            # Save the cropped image with the desired naming convention
            crop_name = f"frame_{frame_count}_ID_{results[0].boxes.id[i]}.jpg"
            crop_path = os.path.join(save_dir, crop_name)
            cv2.imwrite(crop_path, crop)

        frame_ = results[0].plot()

        cv2.imshow("frame", frame_)
        if cv2.waitKey(25) & 0xFF == ord('q'):
            break

cap.release()
cv2.destroyAllWindows()

In this code, we loop through each detection in the current frame, extract the bounding box coordinates, crop the detected object from the frame, and save it with the naming convention you specified. Make sure to replace "path/to/save/crops" with the actual path where you want to save the cropped images.

Please note that this code assumes that the model.track() method returns the IDs in the results[0].boxes.id attribute. If the IDs are not available or the attribute name is different, you'll need to adjust the code accordingly.

I hope this helps! If you have any more questions or need further assistance, feel free to ask. Happy coding! 🚀

[AdadAlShabab]

Hi @pderrenger . Can I run the models using my phones camera? Can you please share the code to invoke my mobile's camera to test the model? Thanks in advance.

[pderrenger]

@AdadAlShabab hi there! 👋 Yes, you can use your phone's camera to test the YOLO models. If you're looking to connect directly from your phone, you might need an app that can stream your camera feed to a URL. Then, you can use that URL as the source in the model.track() method. Here's a basic example of how you might set it up in Python:

from ultralytics import YOLO

# Load your trained model
model = YOLO('path/to/your/model.pt')

# Replace 'your_stream_url' with the URL from your phone's camera stream
results = model.track(source='your_stream_url', show=True)

Make sure your phone and the computer running the model are on the same network. Happy tracking! 😊

[Liuchunxiao111]

There's no problem at all using your way, but when I use it, the video streaming latency is extraordinarily high, why is that?

import cv2
from ultralytics import YOLO

# Load the YOLOv8 model
model= YOLO('weights/det/yolov8n.pt')

# Open the video file
cap = cv2.VideoCapture("http://192.168.9.123:4747/video")

# Loop through the video frames
while cap.isOpened():
    # Read a frame from the video
    success, frame = cap.read()

    if success:
        # Run YOLOv8 tracking on the frame, persisting tracks between frames
        results = model.track(frame, persist=True)

        # Visualize the results on the frame
        annotated_frame = results[0].plot()

        # Display the annotated frame
        cv2.imshow("YOLOv8 Tracking", annotated_frame)

        # Break the loop if 'q' is pressed
        if cv2.waitKey(1) & 0xFF == ord("q"):
            break
    else:
        # Break the loop if the end of the video is reached
        break

# Release the video capture object and close the display window
cap.release()
cv2.destroyAllWindows()

[pderrenger]

Hi there! 🙌 The high latency you're experiencing during video streaming might be due to a few factors. Here are a couple of things you could check or adjust to potentially reduce the latency:

1. Network Speed: If you're streaming the video over a network, ensure that your network connection is fast and stable. Camera streaming apps can sometimes consume a lot of bandwidth.

2. Frame Processing: The model's predictions and the layer of annotations being added to each frame can be computationally intensive. You could try to reduce the image size for detection or lower the frame rate to see if it helps.

3. Hardware Acceleration: If possible, make sure that your system is using any available hardware acceleration to speed up video decoding and model inference.

4. Less Compute-Intensive Model: If latency is still an issue, consider using a smaller, less compute-intensive YOLO model.

Here’s a quick example of reducing image size which might help speed up the process:

import cv2
from ultralytics import YOLO

# Load the YOLOv8 model
model = YOLO('weights/det/yolov8n.pt')

# Reduce frames size to increase performance
def resize_frame(frame, scale=0.5):
    width = int(frame.shape[1] * scale)
    height = int(frame.shape[0] * scale)
    dimensions = (width, height)

    return cv2.resize(frame, dimensions, interpolation=cv2.INTER_AREA)

# Open the video file
cap = cv2.VideoCapture("http://192.168.9.123:4747/video")

# Loop through the video frames
while cap.isOpened():
    success, frame = cap.read()

    if success:
        # Resize frame
        frame = resize_frame(frame, scale=0.5)

        # Run YOLOv8 tracking on the frame
        results = model.track(frame, persist=True)

        # Visualize the results
        annotated_frame = results[0].plot()
        cv2.imshow("YOLOv8 Tracking", annotated_frame)

        if cv2.waitKey(1) & 0xFF == ord("q"):
            break
    else:
        break

cap.release()
cv2.destroyAllWindows()

This adjusts the frame size being processed, which could potentially reduce the latency you're experiencing. Hope this helps! 😊 If issues persist, feel free to provide more information about your setup.

[Liuchunxiao111]

@pderrenger Thanks for your reply!

What puzzles me is that when I use the first method, the picture is smooth and when I use the second method, the picture has a high latency, why is that? I don't think it's a network or hardware acceleration issue.

Way 1

from ultralytics import YOLO

# Load your trained model
model= YOLO('weights/det/yolov8n.pt')

# Replace 'your_stream_url' with the URL from your phone's camera stream
results = model.track(source="http://192.168.9.123:4747/video", show=True)

Way2

import cv2
from ultralytics import YOLO

# Load the YOLOv8 model
model= YOLO('weights/det/yolov8n.pt')

# Open the video file
cap = cv2.VideoCapture("http://192.168.9.123:4747/video")

# Loop through the video frames
while cap.isOpened():
    # Read a frame from the video
    success, frame = cap.read()

    if success:
        # Run YOLOv8 tracking on the frame, persisting tracks between frames
        results = model.track(frame, persist=True)

        # Visualize the results on the frame
        annotated_frame = results[0].plot()

        # Display the annotated frame
        cv2.imshow("YOLOv8 Tracking", annotated_frame)

        # Break the loop if 'q' is pressed
        if cv2.waitKey(1) & 0xFF == ord("q"):
            break
    else:
        # Break the loop if the end of the video is reached
        break

# Release the video capture object and close the display window
cap.release()
cv2.destroyAllWindows()

[pderrenger]

Hey @Liuchunxiao111! 🌟 The difference in latency between the two methods might be due to how the frames are processed and displayed. In the first method, the model.track() function handles both the tracking and the frame retrieval internally, which is optimized for efficiency. In the second method, using OpenCV's VideoCapture and processing each frame in a loop with model.track() introduces overhead, especially if the frame processing isn't keeping up with the frame rate of the video stream.

To reduce latency in the second method, you might consider:

• Reducing the resolution of the video stream.
• Simplifying the model if possible, or using a lighter model.
• Ensuring that the code runs on the most capable hardware available (like using a GPU if available).

These tweaks can help align the processing speed more closely with the incoming frame rate, reducing the noticeable lag. Hope this helps! 😊

[Boaruzhanchik]

Hi, help me understand why I get this error when tracking with segmentation model . My ultimate goal is to use a custom car plate segmentation model for tracking. Thank you very much

  File "/home/iosif/Documents/Dev/yolo_model/yolov8_train/yolov8_tracking_car_number/tracking_v4.py", line 80, in <module>
    annotated_frame = results[0].plot()
  File "/home/iosif/.local/lib/python3.10/site-packages/ultralytics/engine/results.py", line 239, in plot
    annotator.masks(pred_masks.data, colors=[colors(x, True) for x in idx], im_gpu=im_gpu)
  File "/home/iosif/.local/lib/python3.10/site-packages/ultralytics/utils/plotting.py", line 237, in masks
    self.im[:] = im_mask_np if retina_masks else ops.scale_image(im_mask_np, self.im.shape)
ValueError: could not broadcast input array from shape (384,640,3) into shape (568,1012,3)


from ultralytics import YOLO
import cv2
from collections import defaultdict
import numpy as np

class_list = [0, 2]
model = YOLO("/home/iosif/Documents/Dev/yolo_model/yolov8_train/yolov8s-seg.pt")
cap = cv2.VideoCapture('/home/iosif/Documents/Dev/yolo_model/yolov8_train/yolov8_tracking_car_number/reac_high_way2.mp4')
track_history = defaultdict(lambda: [])

while cap.isOpened():
    success, frame = cap.read()

    if success:
        results = model.track(frame, show=False, verbose=False, conf=0.4, classes=class_list, persist=True)
        if results[0].boxes.id is not None:
            boxes = results[0].boxes.xywh.cpu()
            track_ids = results[0].boxes.id.cpu().numpy().astype(int)
            annotated_frame = results[0].plot()

            for box, track_id in zip(boxes, track_ids):
                x, y, w, h = box
                track = track_history[track_id]
                track.append((float(x), float(y)))
                if len(track) > 40:
                    track.pop(0)
                points = np.hstack(track).astype(np.int32).reshape((-1, 1, 2))
                cv2.polylines(annotated_frame, [points], isClosed=False, color=(0, 0, 255), thickness=5)
            cv2.imshow("YOLOv8 Tracking", annotated_frame)


            class_ids = []
            boxes_id = results[0].boxes.id
            if len(boxes_id) > 1:
                id_values = boxes_id.tolist()
            else:
                id_value = boxes_id.item()

            clss = results[0].boxes.cls.cpu().tolist()
            names = results[0].names
            for box, cls in zip(boxes_id, clss):
                class_ids.append(cls)
                if cls == 0.0:
                    print("People detect")

        

        if cv2.waitKey(1) & 0xFF == ord("q"):
            break
    else:
        break

cap.release()
cv2.destroyAllWindows()

[pderrenger]

Hey there! 👋 It looks like you're encountering a shape mismatch error when plotting your tracking results. This often happens when the image dimensions don't align with the mask dimensions expected by the plot() function.

To resolve this, ensure that the image size you're using for tracking matches the size expected by the model. If you're using a custom model, double-check that the image size during training and inference is consistent.

Here's a quick fix you might try:

# Before plotting, resize your frame to match the model's expected input size
frame_resized = cv2.resize(frame, (model.imgsz[1], model.imgsz[0]))

# Now use the resized frame for tracking and plotting
results = model.track(frame_resized, show=False, verbose=False, conf=0.4, classes=class_list, persist=True)

Make sure to replace model.imgsz[1] and model.imgsz[0] with the actual width and height used during training if model.imgsz is not available.

Give this a shot, and if you're still stuck, feel free to share more details or reach out again! Happy tracking! 🚗💨

[Boaruzhanchik]

Thank you more for such a quick response, it worked

[glenn-jocher]

@Boaruzhanchik hey, that's awesome to hear it worked out for you! 🎉 If you dive into tracking with your custom car plate segmentation model and have any more questions or cool results to share, don't hesitate to reach out. Happy tracking! 🚗🔍

[Elvis-Kiilu2]

@glenn-jocher Hey, how do you implement with custom seg modem, kindly illustrate

[glenn-jocher]

Hey there! Implementing tracking with a custom segmentation model is pretty straightforward. You'll want to load your model and then use the track() method, specifying any particular classes you're interested in. Here's a quick example using a custom segmentation model for car plates:

from ultralytics import YOLO

# Load your custom segmentation model
model = YOLO('path/to/your/custom-seg-model.pt')

# Run tracking on your video source
results = model.track(source='path/to/your/video.mp4', classes=[your_car_plate_class_id])

Just replace 'path/to/your/custom-seg-model.pt' with the actual path to your model and your_car_plate_class_id with the class ID for car plates in your model. If you need more help or have specific questions, feel free to ask! 🚀

[potatoouoo]

Yolov8 has very high overall practicality. Can I implement tracking with two cameras? I hope that when a car tracked by camera A moves to camera B, its frame ID remains the same. However, currently there is always an ID switch happening. Is it because of the model's accuracy?
EX:
def cam1():
model = YOLO('updated-model.pt')
# cap.imgCap()
results = model.track(source="http://192.168.50.204:8080/?action=stream",show = True)
cv2.waitKey(0)
cv2.destroyAllWindows()

def cam2():
model = YOLO('updated-model.pt')
# cap.imgCap()
results = model.track(source="http://192.168.50.69:8080/?action=stream",show = True)
cv2.waitKey(0)
cv2.destroyAllWindows()

cap=cam

a = threading.Thread(target=cam1)
b = threading.Thread(target=cam2)

a.start()
b.start()

[glenn-jocher]

Hey there! Tracking continuity across multiple cameras, especially to maintain the same ID as an object moves from camera A to camera B, is indeed a complex task that involves more than just the accuracy of the model. It requires a method to correlate tracks across the different camera views, often referred to as multi-camera tracking or re-identification (Re-ID).

Regarding access to the global track ID or maintaining consistent IDs across cameras, it's correct that basetrack.py (or similar files, depending on the tracking implementation you're using) manages these IDs. However, for multi-camera setups, you might need to implement or use an existing Re-ID system that can match IDs based on features or appearance.

A simplistic approach could involve:

1. Extracting features of tracked objects.
2. Comparing features of new detections in the second camera with ongoing tracks from the first camera.
3. Assigning the same ID based on the similarity of the features.

This can get quite involved, as it requires managing the timing and spatial-correlation of detections across cameras, and the model's accuracy is just one of many factors.

For directly accessing or setting the track ID, if you're using a specific tracker that has its global track ID in basetrack.py, you might interact with it directly within your code (varying by implementation). However, multi-camera ID continuity typically requires a higher-level management system than what's provided in individual tracker files.

I hope that sheds a bit more light on it! Let me know if you delve into the re-identification approach or if you have other questions. 😊👍

[potatoouoo]

I have a new question, can I know the distance to the camera while tracking?

[glenn-jocher]

Hey there!

Regarding your question about knowing the distance to the camera while tracking, YOLOv8 or most object tracking algorithms, including those implemented in Ultralytics YOLO, primarily focus on detecting objects and potentially tracking their movement across frames. They don't inherently provide distance measurements since that would require additional depth-sensing hardware or methods.

However, if you have a calibrated camera setup or additional information about the scene, you might estimate distances using object sizes or other cues. For instance, in specific, controlled environments, the apparent size of known objects can give clues about distance based on their expected real-world size.

Alternatively, employing depth cameras or stereo vision setups can directly provide depth information. If using such sensors, depth data can be combined with tracking data to estimate distances of tracked objects.

I hope that helps a bit! If you're diving deeper into this, consider exploring topics like camera calibration, depth sensing technologies, and stereo vision. 😊📏🎥

[potatoouoo]

Hello, I would like to ask, my current tracking program is as shown below
retList2, frameTmp2 = cap2.imgCap()
results = model.track(frameTmp2,persist=True,show = True)

             if text2:
                 boxes = results[0].boxes.xywh.cpu()
                 if results[0].boxes.id is not None:
                     track_ids1 = results[0].boxes.id.int().cpu().tolist()
                     plate_info = tracker.get_plate_info()
                     print(plate_info)
                     time.sleep(1)

I need the X1X2Y1Y2 of the tracking frame in each frame to calculate the distance. Is there a way to get it?

[glenn-jocher]

Hello! To get the bounding box coordinates (X1, Y1, X2, Y2) for each tracked object, you can access the .xyxy attribute from the results object. Here’s how you can modify your code to print these coordinates along with the track IDs:

def cam1():
    global shard_track
    model = YOLO('updated-model.pt')
    shard_track = model.track(source="http://192.168.50.204:8080/?action=stream", show=True)
    cv2.waitKey(0)
    cv2.destroyAllWindows()

def track_and_print_cam1():
    global shard_track
    if hasattr(shard_track, 'xyxy'):
        for det in shard_track.xyxy[0]:  # Assuming shard_track is the results object
            x1, y1, x2, y2, conf, cls, track_id = det
            print(f'Track ID: {track_id.item()}, BBox: ({x1.item()}, {y1.item()}, {x2.item()}, {y2.item()})')
    else:
        print("No tracking data found.")

This modification ensures you're extracting the bounding box coordinates directly from the detection results, which should help you calculate distances or any other metrics based on the positions of the objects. If you have any more questions or need further assistance, feel free to ask! 😊

[samyak197]

Hey there,
First of all, I love working with YOLO , functionalities for basically everything.

          Currently i was working with a project that requires me to track different employee in a shop and was wandering if YOLO could help me assigning identities to each employees and if i t cpuld track for the entire working day.

[pderrenger]

@samyak197 hey there! 😊

Absolutely, YOLO can assist with tracking employees in a shop. You can use our tracking models to assign unique IDs and track them throughout the day. For long-term tracking, make sure your video feed is continuous and the model is configured for persistent tracking.

Here's a quick example using YOLOv8:

from ultralytics import YOLO

# Load your custom trained model or an official YOLOv8 model
model = YOLO('path/to/your/custom-model.pt')  # Custom trained model

# Track employees in a video
results = model.track(source="path/to/shop/video.mp4", persist=True)

This will keep the identities consistent across frames. For more details, check out our tracking documentation. If you have any more questions or need further assistance, feel free to ask. Happy tracking! 🚀

[ParkerLischwe]

Hi I saw that I can use an openvino IR format model just like any other pytorch model and then run tracking like normal. I was wondering how I would load the IR '.xml' and '.bin' files as arguments into YOLO(), or if I should load my model using openvino library? Thanks.

[glenn-jocher]

Hey there! 👋

To load OpenVINO IR '.xml' and '.bin' files into YOLO(), you'll want to simply point to their containing directory, i.e.:

from ultralytics import YOLO

# Load your YOLOv8 OpenVINO model
model = YOLO('path/to/yolov8n_openvino_model/')  # exported with YOLO('yolov8n.pt').export(format='openvino')

# Track employees in a video
results = model.track(source="path/to/shop/video.mp4", persist=True)

Once you have your model loaded with OpenVINO, you can then proceed with tracking as you would normally with the YOLO model. You might want to use stream=True also for longer videos and streams. If you need further assistance, feel free to ask. Happy tracking! 😊

[ParkerLischwe]

Will give this a try, appreciate the timely response!

[pderrenger]

@ParkerLischwe glad to hear you're giving it a try! If you have any questions or need a hand with the tracking setup, just drop a message. Happy tracking! 😊🚀

[ANIRUDH1080Z]

Can i use yolov8 model to track and reidentify person with same id assigned to it in multiple camera feed ?

[glenn-jocher]

@ANIRUDH1080Z yes, you can use YOLOv8 models for multi-object tracking across multiple camera feeds. Each detected person can be assigned a unique ID that persists as they move through the field of view of different cameras. You can choose from different trackers like BoT-SORT and ByteTrack, and even customize the tracker configurations to suit your specific needs. Here's a simple example using the default tracker:

from ultralytics import YOLO

# Load a YOLOv8 model
model = YOLO('yolov8n.pt')

# Perform tracking on a video source
results = model.track(source="path/to/camera_feed.mp4", show=True)

For tracking across multiple cameras, you might need to synchronize the cameras and handle the ID reidentification logic, as the trackers maintain IDs within a single video stream. If you have a custom solution or improvements, consider contributing to our Trackers section! 🚀

[ANIRUDH1080Z]

Thank u so much for enlightening me about this.

[glenn-jocher]

You're welcome! I'm glad the information was helpful. If you're looking to implement multi-object tracking across multiple camera feeds, you can certainly use YOLOv8 models. While YOLOv8 can assign unique IDs within a single video stream, tracking across multiple cameras (especially with non-overlapping fields of view) typically requires additional logic to handle ID reidentification.

For such cases, you might consider integrating your YOLOv8 tracking with a reidentification (ReID) system that can match IDs based on appearance features. Here's a simple example of how you might set up tracking for a single camera feed:

from ultralytics import YOLO

# Load a YOLOv8 model
model = YOLO('yolov8n.pt')

# Perform tracking on a video source
results = model.track(source="path/to/camera_feed.mp4", show=True)

For multi-camera setups, after tracking, you would extract features from detected objects and use a ReID model to match identities across different camera feeds.

If you're developing a solution or have improvements, we'd love to see your contributions to our tracking capabilities! 🚀

[kshweikeh]

How can we only track moving objects in the Plotting Tracks Over Time code:

from collections import defaultdict

import cv2
import numpy as np

from ultralytics import YOLO

Load the YOLOv8 model

model = YOLO('yolov8n.pt')

Open the video file

video_path = "path/to/video.mp4"
cap = cv2.VideoCapture(video_path)

Store the track history

track_history = defaultdict(lambda: [])

Loop through the video frames

while cap.isOpened():
# Read a frame from the video
success, frame = cap.read()

if success:
    # Run YOLOv8 tracking on the frame, persisting tracks between frames
    results = model.track(frame, persist=True)

    # Get the boxes and track IDs
    boxes = results[0].boxes.xywh.cpu()
    track_ids = results[0].boxes.id.int().cpu().tolist()

    # Visualize the results on the frame
    annotated_frame = results[0].plot()

    # Plot the tracks
    for box, track_id in zip(boxes, track_ids):
        x, y, w, h = box
        track = track_history[track_id]
        track.append((float(x), float(y)))  # x, y center point
        if len(track) > 30:  # retain 90 tracks for 90 frames
            track.pop(0)

        # Draw the tracking lines
        points = np.hstack(track).astype(np.int32).reshape((-1, 1, 2))
        cv2.polylines(annotated_frame, [points], isClosed=False, color=(230, 230, 230), thickness=10)

    # Display the annotated frame
    cv2.imshow("YOLOv8 Tracking", annotated_frame)

    # Break the loop if 'q' is pressed
    if cv2.waitKey(1) & 0xFF == ord("q"):
        break
else:
    # Break the loop if the end of the video is reached
    break

Release the video capture object and close the display window

cap.release()
cv2.destroyAllWindows()

[kshweikeh]

I forgot to write "only " so its MOVING OBJECTS ONLY

[glenn-jocher]

@kshweikeh hey there! 👋 To track only moving objects, you can modify the code to calculate the displacement of each object between frames and only plot the tracks for those with significant movement. Here's a quick example of how you might do that:

# ... [previous code]

# Define a movement threshold (this value may need tuning)
movement_threshold = 5  # pixels

# Loop through the video frames
while cap.isOpened():
    # ... [previous code]

    # Get the previous and current center points
    for track_id in track_ids:
        if len(track_history[track_id]) >= 2:
            prev_center = track_history[track_id][-2]
            curr_center = track_history[track_id][-1]
            displacement = np.sqrt((curr_center[0] - prev_center[0]) ** 2 + (curr_center[1] - prev_center[1]) ** 2)

            # Only plot the track if the displacement is above the threshold
            if displacement > movement_threshold:
                points = np.array(track_history[track_id], dtype=np.int32).reshape((-1, 1, 2))
                cv2.polylines(annotated_frame, [points], isClosed=False, color=(230, 230, 230), thickness=10)

    # ... [previous code]

This snippet calculates the Euclidean distance between the previous and current positions of each tracked object and only draws the track if the object has moved more than the movement_threshold. You can adjust the threshold to fit the specifics of your scenario. Happy coding! 🚀

[kshweikeh]

Thanks!!

[glenn-jocher]

@kshweikeh hey there! 👋

Absolutely, happy to help! For tracking moving objects with YOLOv8, you can utilize the track method with a custom tracker configuration that focuses on objects with significant displacement between frames. Here's a quick snippet to get you started:

from ultralytics import YOLO

# Load your model
model = YOLO('path/to/your/model.pt')

# Run the tracker
results = model.track(source='path/to/your/video.mp4', tracker='custom_tracker.yaml')

In your custom_tracker.yaml, you might want to adjust parameters that affect the detection of stationary vs. moving objects, such as minimum displacement thresholds or object persistence criteria.

If you need more guidance or have specific requirements, feel free to share, and we can dive deeper! 🚀

[RNVALA]

import cv2
from ultralytics import YOLO
import numpy as np

model = YOLO('yolov8_custom_train.engine', task="detect")

Path to the input video file

input_video_path = '/content/gdrive/MyDrive/yolov8-tensorrt/inference/output_video.mp4'

Path to the output video file

output_video_path = 'outputtest_video.mp4'

Define the coordinates of the polygon

polygon_points = [(670, 66), (1237, 550), (514, 1054), (161, 295)]
polygon_points = np.array(polygon_points)

Open the input video file

cap = cv2.VideoCapture(input_video_path)

Get video properties

frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
fps = int(cap.get(cv2.CAP_PROP_FPS))
total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))

Define the codec and create VideoWriter object

fourcc = cv2.VideoWriter_fourcc(*'mp4v')
out = cv2.VideoWriter(output_video_path, fourcc, fps, (frame_width, frame_height))

Function for finding the centroid

def calculate_centroid(box):
x_min, y_min, x_max, y_max = map(int, box)
centroid_x = (x_min + x_max) // 2
centroid_y = (y_min + y_max) // 2
return (centroid_x, centroid_y)

Function to check if two bounding boxes overlap

def check_overlap(box1, box2):
x1_min, y1_min, x1_max, y1_max = map(int, box1)
x2_min, y2_min, x2_max, y2_max = map(int, box2)
overlap_x = max(0, min(x1_max, x2_max) - max(x1_min, x2_min))
overlap_y = max(0, min(y1_max, y2_max) - max(y1_min, y2_min))
overlap_area = overlap_x * overlap_y
return overlap_area

Read until video is completed

while cap.isOpened():
# Read a frame from the video
ret, frame = cap.read()

# If frame is read correctly ret is True
if not ret:
    break

results = model.track(source=frame, device=0)
# Extract bounding boxes, classes, names, and confidences
boxes = results[0].boxes.xyxy.tolist()
classes = results[0].boxes.cls.tolist()
names = results[0].names
confidences = results[0].boxes.conf.tolist()
#id=results[0].boxes.id
# Iterate through the results
red_area_count = 0
blue_area_count = 0
crowded_area_detected = False
crowded_box = None

for i, box1 in enumerate(boxes):
    overlap_count = 0
    for j, box2 in enumerate(boxes):
        if i != j:
            overlap_area = check_overlap(box1, box2)
            if overlap_area > 0:
                overlap_count += 1
                if overlap_count >= 2:
                    crowded_area_detected = True
                    if crowded_box is None:
                        crowded_box = box1
                    else:
                        crowded_box = [
                            min(crowded_box[0], box1[0]),
                            min(crowded_box[1], box1[1]),
                            max(crowded_box[2], box1[2]),
                            max(crowded_box[3], box1[3])
                        ]
                    break

for box, cls, conf in zip(boxes, classes, confidences):
    x1, y1, x2, y2 = box
    confidence = conf
    detected_class = cls
    name = names[int(cls)]
    
    if cv2.pointPolygonTest(polygon_points, calculate_centroid(box), False) >= 0:
        red_area_count += 1
        color = (0, 0, 255)
    else:
        color = (0, 255, 0)
        blue_area_count += 1

    x1, y1, x2, y2 = int(x1), int(y1), int(x2), int(y2)
    cv2.rectangle(frame, (x1, y1), (x2, y2), color, 2)
for i,det in enumerate(results[0].boxes.xyxy):
    x1, y1, x2, y2 = map(int, det[:4])
    cv2.putText(frame, f'ID: {results[0].boxes.id[i]}', (x1+5, y1+25), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (255,255,255), 2, cv2.LINE_AA)
    
if crowded_area_detected:
    crowded_boxes = [box for box in boxes if check_overlap(crowded_box, box) > 0]
    x1, y1, x2, y2 = crowded_box
    for box in crowded_boxes:
        if box != crowded_box and check_overlap(crowded_box, box) > 0:
            x1 = min(x1, box[0])
            y1 = min(y1, box[1])
            x2 = max(x2, box[2])
            y2 = max(y2, box[3])
    cv2.rectangle(frame, (int(x1), int(y1)), (int(x2), int(y2)), (0, 255, 255), 2)
    cv2.putText(frame, 'Crowded Area', (int(x1), int(y1) - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 255, 255), 2)

font = cv2.FONT_HERSHEY_SIMPLEX

cv2.putText(frame, 'People working with attention - {}'.format(blue_area_count), (1125, 24), font, 1, (0, 255, 0), 2, cv2.LINE_AA)
cv2.putText(frame, 'People not working - {}'.format(red_area_count), (1124, 75), font, 1, (0, 0, 255), 2, cv2.LINE_AA)
cv2.line(frame, (668, 66), (1237, 550), (255, 0, 0), 2)
cv2.line(frame, (1237, 550), (515, 1052), (255, 0, 0), 2)

# Write the frame into the output video file
out.write(frame)

# Display the frame
# cv2_imshow(frame)

# Press Q on keyboard to exit
if cv2.waitKey(25) & 0xFF == ord('q'):
    break

Release video objects

cap.release()
out.release()

Close all OpenCV windows

cv2.destroyAllWindows()

In this, I am tracking a label Person but in the next 2 to 3 frames, ids are changing so any solution for this?
means if person having id 1, so after few frame it give as id 3, and every time the id is changing

[pderrenger]

Hey there! 👋 It looks like you're experiencing some ID consistency issues with your custom-trained model during tracking. This can happen due to various reasons, such as similar-looking objects or occlusions.

One approach to mitigate this is to fine-tune your tracker's hyperparameters. If you're using the default BoT-SORT tracker, you might want to experiment with the bytetrack.yaml configuration, which is designed to handle ID switches more robustly. Here's a quick example of how to switch to ByteTrack:

from ultralytics import YOLO

# Load your custom-trained model
model = YOLO('yolov8_custom_train.engine', task="detect")

# Perform tracking with ByteTrack
results = model.track(source="your_video.mp4", tracker="bytetrack.yaml")

If you've already tried adjusting the tracker settings and the issue persists, consider revisiting your training dataset to ensure it's diverse and representative enough to handle the scenarios where ID switches occur.

Keep us posted on your progress, and if you need more assistance, we're here to help! 😊

[RNVALA]

I have already used the bytetrack.yaml and this is ouput video , https://drive.google.com/file/d/1E95m456dbZAUDRrodfrvrsSA4zcs3KiA/view?usp=sharing , please analyze it

[pderrenger]

Hey there! Thanks for reaching out and sharing the video link. 🎥 I took a look, and it seems like the ID switches might be due to occlusions or similar-looking objects. Here are a couple of quick suggestions:

1. Tweak the IOU Threshold: Adjust the iou_thresh parameter in your bytetrack.yaml to a value that might help in maintaining consistent IDs.
2. Increase Detection Confidence: A higher conf_thresh could reduce false positives, which sometimes cause ID confusion.

If you've already fine-tuned these without success, consider:

• Re-training with Augmented Data: More varied training data can improve tracking robustness.
• Post-Processing: Implement a custom ID management system in post-processing to smooth out ID switches.

If you're still facing issues, feel free to share more details or reach out for further assistance. We're here to help! 😊👍

[Dmonzer]

What is the difference between these attributes of results[0].boxes:
xywh: tensor([], size=(0, 4))
xywhn: tensor([], size=(0, 4))
xyxy: tensor([], size=(0, 4))
xyxyn: tensor([], size=(0, 4))

[glenn-jocher]

Hey there! 👋 The attributes you're asking about represent different formats of bounding box coordinates:

• xywh: Bounding boxes in (center x, center y, width, height) format.
• xywhn: Normalized (center x, center y, width, height), with values from 0 to 1 relative to image size.
• xyxy: Bounding boxes in (top left x, top left y, bottom right x, bottom right y) format.
• xyxyn: Normalized (top left x, top left y, bottom right x, bottom right y), with values from 0 to 1.

The tensors being empty (size=(0, 4)) suggests no detections were made in the frame. If you have further questions or need assistance, feel free to ask! Happy coding! 😊

[WooFy11]

is it possible to use our own weighs as a model to track? or we must include the yolov8n.pt?

[pderrenger]

Absolutely, you can use your own custom weights for tracking! Just load your trained model instead of the default yolov8n.pt. Here's a quick example in Python:

from ultralytics import YOLO

# Load your custom trained model
model = YOLO('path/to/your/custom-weights.pt')

# Perform tracking with your model
results = model.track(source="path/to/your/video.mp4", show=True)

And that's it! Your custom model will now be used for tracking objects in your video stream. 🚀 If you have any more questions or need further assistance, feel free to ask. Happy tracking!

[WooFy11]

do you a discord so we can discuss about my project?

[pderrenger]

Hey there! 👋 We don't have an official Discord channel, but we're happy to help you right here or on our GitHub discussions page. If you have any questions about your project or need assistance with object tracking using YOLOv8, feel free to drop your questions, and we'll do our best to assist you! 🚀 If it's code-specific, you can also share snippets or error messages, and we'll take a look. Happy tracking!

[WooFy11]

i had an error when everytime i try to run a code, it crashed on 4xx/600, and i must restart it, and 2nd question can we make a custom annotator?

[pderrenger]

Hey there! 👋 It sounds like you're encountering a crash during your code execution. To better assist you, could you please share the specific error message you're getting? This will help us pinpoint the issue more accurately.

Regarding your second question, yes, you can create a custom annotator. If you're working with YOLOv8, you can use tools like CVAT or LabelImg for annotation, and then train your model with these custom annotations. Here's a quick example of how you might load and use your custom model for tracking:

from ultralytics import YOLO

# Load your custom trained model
model = YOLO('path/to/your/custom_model.pt')

# Perform tracking on your video
results = model.track(source='path/to/your/video.mp4', show=True)

This will use your custom weights for tracking objects in the video. If you have any more questions or need further assistance, feel free to ask! 🚀

[tumitoto007]

So I am using Yolov8 for my current project, it's been a breeze so far. I do have a question on the tracking method provided by the Yolov8. When I am using the generic yolov8n model(or even a custom model mixed with few objects), I know I can specifically filter out things that doesn't interest me by their ID as below:

    for data in .....boxes.data.tolist():
            # extract the id associated with the detection
            objectID= data[5]
   if not interested skip this object....

But, when I caught the an object that I am interested, can I at that time or at that frame, issue a track command to start tracking it? if it can be done, can you tell me how? an short example will be even better! thanks in advance.

[glenn-jocher]

@tumitoto007 hey there! 😊 It's great to hear that YOLOv8 has been a breeze for your project. Regarding your question about initiating tracking on a specific object at a certain frame, YOLOv8's tracking functionality is designed to automatically track all detected objects continuously across frames. However, if you're interested in tracking a specific object after its detection, you can filter detections by their IDs as you've mentioned and then focus on tracking that specific object ID across the frames.

Here's a simplified example to illustrate how you might approach this:

from ultralytics import YOLO

# Load your model
model = YOLO('yolov8n.pt')  # or your custom model

# Assuming you're processing video frames in a loop
interested_object_id = None  # Placeholder for the ID of the object you're interested in tracking
for frame in video_frames:
    results = model.track(frame, persist=True)
    for box in results[0].boxes:
        objectID = box.id.item()  # Get the ID of the current object
        if objectID == interested_object_id:
            # This is the object you're interested in, now you can track it specifically
            # Process the tracking data as needed
            pass

In this example, interested_object_id should be set to the ID of the object you're interested in tracking once you've identified it. From that point on, you can specifically process or track that object in subsequent frames.

Remember, the tracking starts automatically for all detected objects, so the key is filtering and focusing on the object of interest in your processing loop.

Hope this helps! If you have any more questions or need further assistance, feel free to ask. Happy tracking! 🚀

[tumitoto007]

Cool! I will give this a try once I get back to town. Thanks again for your great support.

[glenn-jocher]

Hey there! 🎉 I'm thrilled to hear you're planning to give tracking a try with YOLOv8 once you're back in town. It sounds like you're on an exciting journey with your project! If you have any questions or need further assistance when you dive into it, don't hesitate to reach out. Our community and docs are here to support you every step of the way. Happy tracking, and safe travels back to town! 🚀

[KhizarHashmi]

Hi, I want some detailed help and guidance on how to use custom tracker models with my custom yolov8 pose model, the Re-identification problem is being face using bytetrack.yaml so I think I should use StrongSORT or DeepSORT. Therefore, I want the ultralytics team to help me on selecting my tracker model or use multiple tracker models, and guide me properly on how to use them with my YOLOv8 custom trained model.
Thanks

[KhizarHashmi]

@glenn-jocher

[glenn-jocher]

@KhizarHashmi hi there! 👋 It sounds like you're exploring advanced tracking options for your custom YOLOv8 pose model. If ByteTrack isn't meeting your re-identification needs, exploring StrongSORT or DeepSORT could indeed be beneficial. These trackers offer sophisticated mechanisms for handling occlusions and maintaining identity across frames, which might be what you're looking for.

To integrate a custom tracker like StrongSORT or DeepSORT with your YOLOv8 model, you'll typically need to:

1. Ensure your tracker can process the detection output format of YOLOv8. This usually involves bounding box coordinates, confidence scores, and class IDs.
2. Modify your tracking script to use the YOLOv8 model for detection, feeding these detections into your chosen tracker for each frame of your video.

Here's a simplified example of how you might set this up in Python, assuming you have a function custom_tracker that takes detections and returns tracked objects:

from ultralytics import YOLO

# Load your custom trained YOLOv8 model
model = YOLO('path/to/your/custom_model.pt')

# Function to perform detection and tracking on a frame
def detect_and_track(frame):
    # Perform detection
    results = model(frame)
    
    # Extract detections (example format, adjust as needed)
    detections = results.pandas().xyxy[0]  # DataFrame of detections
    
    # Perform tracking with your custom tracker
    tracked_objects = custom_tracker(detections)
    
    return tracked_objects

# Example usage with a video stream
# ... video capture setup here ...
while True:
    ret, frame = cap.read()
    if not ret:
        break
    tracked_objects = detect_and_track(frame)
    # ... visualization and further processing here ...

Regarding using multiple trackers, it's possible to switch between trackers based on the scenario or even combine their strengths, though this can get complex. You might start with one tracker and evaluate its performance thoroughly before experimenting with combining them.

For more detailed guidance on integrating specific trackers or handling complex tracking scenarios, the documentation and community forums for those trackers can be valuable resources. Additionally, our Ultralytics Docs provide further insights into tracking configurations and options.

If you're developing new tracking capabilities or have insights to share, we'd love to see your contributions to our GitHub repo. Your expertise could greatly benefit the YOLO and tracking community! 🚀

Hope this helps! Let us know if you have more questions.

[Vinrub]

import random
import cv2
import numpy as np
from ultralytics import YOLO

opening the file in read mode

my_file = open("utils/coco.txt", "r")

reading the file

data = my_file.read()

replacing end splitting the text | when newline ('\n') is seen.

class_list = data.split("\n")
my_file.close()

Generate random colors for class list

detection_colors = []
for i in range(len(class_list)):
r = random.randint(0, 255)
g = random.randint(0, 255)
b = random.randint(0, 255)
detection_colors.append((b, g, r))

load a pretrained YOLOv8n model

model = YOLO("weights/yolov8n.pt", "v8")

Vals to resize video frames | small frame optimise the run

frame_wid = 640
frame_hyt = 480

def CarBehaviour(frame, color_threshold=1100):
# Convert the frame to HSV for color thresholding
hsv_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)

# Define the lower and upper bounds for the color of the rear lights (adjust as needed)
lower_color = np.array([0, 100, 100])  # Lower bound for red color
upper_color = np.array([10, 255, 255]) # Upper bound for red color

# Create a mask using the inRange function to threshold the color
mask = cv2.inRange(hsv_frame, lower_color, upper_color)

# Calculate the number of nonzero pixels in the mask
color_pixels = cv2.countNonZero(mask)

# Check if color pixels are present
if color_pixels > color_threshold:
    return True, 'Braking'  # If color threshold is met, indicate braking
else:
    return False, 'Not Braking'  # Otherwise, indicate not braking

def detect_and_draw(frame, model, class_list, detection_colors):
# Predict on image
detect_params = model.track(source=[frame], conf=0.45, save=False)

# Convert tensor array to numpy
DP = detect_params[0].cpu().numpy()

if len(DP) != 0:
    for i in range(len(detect_params[0])):
        boxes = detect_params[0].boxes
        box = boxes[i]  # returns one box
        clsID = box.cls.cpu().numpy()[0]
        conf = box.conf.cpu().numpy()[0]
        bb = box.xyxy.cpu().numpy()[0]

        cv2.rectangle(
            frame,
            (int(bb[0]), int(bb[1])),
            (int(bb[2]), int(bb[3])),
            detection_colors[int(clsID)],
            3,
        )

        # Display class name and confidence
        font = cv2.FONT_HERSHEY_COMPLEX
        cv2.putText(
            frame,
            class_list[int(clsID)] + " " + str(round(conf, 3)) + "%",
            (int(bb[0]), int(bb[1]) - 10),
            font,
            1,
            (255, 255, 255),
            2,
        )

# Check car behavior
is_braking, behavior = CarBehaviour(frame)
# Display car behavior
cv2.putText(
    frame,
    "Car Behavior: " + behavior,
    (50, 50),
    cv2.FONT_HERSHEY_SIMPLEX,
    1,
    (0, 255, 0),
    2,
    cv2.LINE_AA,
)

Open video capture

cap = cv2.VideoCapture("/home/opencv_env/Vehicle-rear-lights-analyser-master/testing_data/road_2.mp4")

if not cap.isOpened():
print("Cannot open camera")
exit()

while True:
# Capture frame-by-frame
ret, frame = cap.read()
# if frame is read correctly ret is True

if not ret:
    print("Can't receive frame (stream end?). Exiting ...")
    break

#  resize the frame | small frame optimise the run
# frame = cv2.resize(frame, (frame_wid, frame_hyt))

# Detect and draw objects
detect_and_draw(frame, model, class_list, detection_colors)

# Display the resulting frame
cv2.imshow("ObjectDetection", frame)

# Terminate run when "Q" pressed
if cv2.waitKey(1) == ord("q"):
    break

When everything done, release the capture

cap.release()
cv2.destroyAllWindows()
in this code can i add tracking id number using yolov8

[glenn-jocher]

Absolutely, you can add tracking IDs to your detections using YOLOv8! 🚀 To do this, you'll want to utilize the track method available in the Ultralytics YOLO API. This method not only performs object detection but also assigns unique IDs to each detected object, allowing you to track their movement across frames.

Here's a simplified example to get you started:

from ultralytics import YOLO

# Load your model
model = YOLO("weights/yolov8n.pt")

# Assuming 'frame' is your input frame
results = model.track(source=[frame])

# Now, each detected object in 'results' has a unique tracking ID
for det in results[0].boxes:
    # Extract tracking ID
    track_id = det.id.cpu().numpy()
    print(f"Tracking ID: {track_id}")

This will print the tracking ID for each detected object in the frame. Make sure to loop over your video frames to continuously track objects. Happy tracking! 🎉

[kush2022]

I have been trying to fine tune my model to track and count fish fingerlings, keep in mind of their small size. Any suggestions of how I can go about this

[glenn-jocher]

@kush2022 to track and count small fish fingerlings, consider using a high-resolution model and adjusting the confidence threshold to better detect smaller objects. You might also explore using a custom tracker configuration to fine-tune the tracking parameters. For more details, please refer to the tracking section in the Ultralytics documentation.

[Aceaks]

The functionality and applications are great of tracking but it would be better to have a functionality to use deep sort, strong sort, boost track++ etc directly, model files we will take care of that wouldn't be an issue. Also i don't know how do i start to integrate any official traking approach like boost track++ with yolo v8 as there is no proper integrated repository and i want to know the process of integrating as i believe it would be better to know rather than looking and finding no solution elsewhere

[glenn-jocher]

@Aceaks thank you for your interest in integrating additional trackers with YOLOv8. Currently, Ultralytics supports BoT-SORT and ByteTrack. For integrating other trackers like Deep SORT or Strong SORT, you'll need to adapt their configurations to work with YOLOv8's output. You can start by exploring the existing tracker configurations in the ultralytics/cfg/trackers directory as a reference. If you have further questions, feel free to ask here.

[Aceaks]

Thanks, that could be nice and unique way

[glenn-jocher]

Thank you for your suggestion. Currently, Ultralytics supports BoT-SORT and ByteTrack. To integrate other trackers like Deep SORT, you'll need to adapt their configurations to work with YOLOv8's output. You can start by exploring the existing tracker configurations in the ultralytics/cfg/trackers directory as a reference. If you have further questions, feel free to ask here.

[Ranjith3436]

how to run a infernece in multiple thread using yolov8 model. i have multipe cpu server. i want to reduce the processing time. i want to use use a multiple cpus for a single inferece . could you help me guys?

[Ranjith3436]

how help this line for this problem..

torch.set_num_threads(6)

also above lines of code how help model.track(frame, device='cpu')

[glenn-jocher]

@Ranjith3436 to run inference using multiple CPUs, you can utilize Python's threading module to parallelize the process. Each thread can handle a separate inference task, allowing you to leverage multiple CPU cores. Here's a minimal example:

import threading
from ultralytics import YOLO

def run_inference(model_path, image_path):
    model = YOLO(model_path)
    results = model(image_path)
    print(results)

model_path = "yolov8n.pt"
image_paths = ["image1.jpg", "image2.jpg"]  # Add more images as needed

threads = []
for image_path in image_paths:
    thread = threading.Thread(target=run_inference, args=(model_path, image_path))
    threads.append(thread)
    thread.start()

for thread in threads:
    thread.join()

This setup allows each thread to perform inference on a different image, effectively utilizing multiple CPUs.

[Ranjith3436]

Hii @glenn-jocher ,
i am not mentioned this problem. i looked it in your docs. i want to handle internally process threads.

torch.set_num_threads(6)

using this set of codes. i want to know

1. is this possible to control processing threads?

2. if it is possible, how can we handle?

3. how many thread used for a single model single video inference?

4. if the load increases, how model take multiple cpus specially high quality footage ?

5. if the load increases some process slow down becasue of minimum amount of CPUs? if we increases the CPUs can be reduced the processing time?

6. then how to handle number of CPU usage for single inference?

[glenn-jocher]

To utilize multiple CPUs for inference, torch.set_num_threads(6) can help control the number of threads PyTorch uses, potentially improving performance. However, for leveraging multiple CPUs effectively, consider using Python's threading or multiprocessing modules to parallelize tasks. Each thread or process can handle separate inference tasks, allowing better CPU utilization. If you need further assistance, please refer to the Ultralytics documentation for more details.

[tastafur]

Hello, I have a problem that I have been thinking about for several days, I have this script that captures all the frames of a streaming camera, what happens is that I infer the video with tracking and the frames come from a queue to capture them before but the tracker is much slower in processing than the videos it captures, then the queue grows until I run out of memory or grows to a maximum and I lose frames and I would like not to lose frames and go processing them all but I can't find a way.

I pass you the example of the implemented code to see if you can give me a hand.

#!/usr/bin/env python3
import cv2
import multiprocessing as mp
import logging
from collections import defaultdict, Counter
from ultralytics import YOLO
import numpy as np
import argparse
from pymongo import MongoClient
import clickhouse_connect
from ultralytics.utils.plotting import Annotator, Colors
from ultralytics.utils.ops import xyxy2ltwh
from dateutil.parser import isoparse
from datetime import datetime, timedelta
from bson import ObjectId
from utils import cam_to_adr, cam_to_direction, day_of_week_translation
import signal
import sys

# Logger configuration
logging.basicConfig(level=logging.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s')

# Output video configuration
output_video_writer = None

# Global variables
track_history = defaultdict(lambda: [])
track_class_history = defaultdict(list)
last_seen_frame = {}
current_class = {}
object_count = defaultdict(int)
frames_until_disappear = 100
grace_period_seconds = 5
current_frame = mp.Value('i', 0)  # Shared value between processes
first_detection_time = {}

# Database variables
mongo_client = None
clickhouse_client = None

# Input video variables
video_codec = None
video_framerate = 30
video_resolution = None

DEFAULT_FPS = 30

# Global event to stop processes
stop_event = mp.Event()

def signal_handler(sig, frame, processes):
    """Signal handler to stop the processes gracefully."""
    logging.info("Interrupt signal received, stopping processes...")
    stop_event.set()  # Stop all processes by setting the event.

    # Try to wait for processes to finish
    for process in processes:
        if process.is_alive():
            logging.info(f"Waiting for process {process.name} to finish...")
            process.join(timeout=5)  # Wait up to 5 seconds for each process

            # If the process is still alive after 5 seconds, terminate it forcefully
            if process.is_alive():
                logging.warning(f"Forcing termination of process {process.name}...")
                process.terminate()
                process.join()

    logging.info("All processes have been stopped successfully.")

    # Release video resources
    if output_video_writer and output_video_writer.isOpened():
        output_video_writer.release()
        logging.info(f"Video saved and closed successfully.")

    # Close MongoDB and ClickHouse connections if open
    if mongo_client:
        mongo_client.close()
    if clickhouse_client:
        clickhouse_client.close()

    cv2.destroyAllWindows()
    sys.exit(0)

def capture_frames(video_source, frame_queue, max_queue_size):
    """Function to capture frames and add them to the queue"""
    logging.info("Starting frame capture...")
    cap = cv2.VideoCapture(video_source)

    global video_framerate, video_resolution, video_codec
    video_codec = cap.get(cv2.CAP_PROP_FOURCC)
    video_framerate = cap.get(cv2.CAP_PROP_FPS) or DEFAULT_FPS

    frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
    frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
    video_resolution = {"width": frame_width, "height": frame_height}

    while cap.isOpened() and not stop_event.is_set():
        ret, frame = cap.read()
        if not ret:
            logging.warning("End of video or error in capture.")
            break

        # Wait until there is space in the queue to add the next frame
        while frame_queue.qsize() >= max_queue_size and not stop_event.is_set():
            # Actively wait until there is space
            continue

        # Add the frame to the processing queue
        if not stop_event.is_set():
            frame_queue.put(frame)
            logging.debug(f"Frame captured and added to queue. Current queue size: {frame_queue.qsize()}")

    cap.release()
    logging.info("Frame capture finished.")


def inference_worker(frame, model):
    """Function to perform inference on a frame"""
    logging.debug("Starting inference on frame.")
    return model.track(frame, conf=0.3, iou=0.5, persist=True, show=False, stream=True, tracker="botsort.yaml")

def process_disappeared_tracks():
    """Check and process disappeared tracks."""
    current_time = datetime.now()
    disappeared_ids = []

    for track_id, (last_seen_frame_num, last_seen_time) in last_seen_frame.items():
        time_since_last_seen = (current_time - last_seen_time).total_seconds()
        frames_since_last_seen = current_frame.value - last_seen_frame_num

        if time_since_last_seen > grace_period_seconds and frames_since_last_seen > frames_until_disappear:
            disappeared_ids.append(track_id)

    for track_id in disappeared_ids:
        predominant_class = Counter(track_class_history[track_id]).most_common(1)[0][0]
        initial_class = current_class[track_id]

        if initial_class != predominant_class:
            object_count[initial_class] -= 1
            object_count[predominant_class] += 1
            logging.info(f"Track ID {track_id} disappeared. Changed class from {initial_class} to {predominant_class}. Count updated.")

        if not args.no_sync:
            save_to_database(track_id, predominant_class)

        del last_seen_frame[track_id]
        del track_class_history[track_id]
        del track_history[track_id]
        del current_class[track_id]
        del first_detection_time[track_id]

def save_to_database(track_id, predominant_class):
    """The code has no influence"""
def process_frames_sequentially(frame_queue, display_queue, model):
    """Function to process frames sequentially and delegate inference"""
    logging.info("Starting sequential frame processing...")
    global current_frame

    colors = Colors() if args.show else None

    while not stop_event.is_set() or not frame_queue.empty():
        try:
            frame = frame_queue.get()
            logging.debug("Frame retrieved from the queue for processing.")
        except mp.queues.Empty:
            continue  # If the queue is empty, wait

        with current_frame.get_lock():
            current_frame.value += 1

        logging.debug(f"Processing frame {current_frame.value}...")

        # Perform inference on the frame
        results = inference_worker(frame, model)

        annotator = Annotator(frame, line_width=3, font_size=16)
        detected_ids = set()

        for result in results:
            if result.boxes is not None and result.boxes.id is not None:
                boxes = result.boxes.xyxy.cpu()
                track_ids = result.boxes.id.int().cpu().tolist()
                classes = result.boxes.cls.int().cpu().tolist()
                scores = result.boxes.conf.cpu().tolist()

                for box, track_id, cls, score in zip(boxes, track_ids, classes, scores):
                    x1, y1, x2, y2 = map(int, box)
                    detected_ids.add(track_id)
                    last_seen_frame[track_id] = (current_frame.value, datetime.now())

                    if track_id not in current_class:
                        initial_class = cls
                        current_class[track_id] = initial_class
                        track_class_history[track_id].append(initial_class)
                        object_count[initial_class] += 1
                        first_detection_time[track_id] = datetime.now()
                        logging.info(f"New track assigned: ID {track_id}, class {initial_class}. Total objects: {object_count[initial_class]}")

                    track_history[track_id].append([x1, y1, x2, y2])
                    if len(track_history[track_id]) > 30:
                        track_history[track_id].pop(0)

                    track_class_history[track_id].append(cls)

                    color = colors(cls)
                    if len(track_history[track_id]) > 1:
                        points = np.array([[(int((b[0] + b[2]) / 2), int((b[1] + b[3]) / 2))] for b in track_history[track_id]]).astype(np.int32)
                        cv2.polylines(annotator.im, [points], isClosed=False, color=color, thickness=2)

                    class_name = model.names[cls] if model.names else f'cls {cls}'
                    label = f'id:{track_id} {class_name} {score:.2f}'
                    annotator.box_label([x1, y1, x2, y2], label, color=color)

        # Process disappeared tracks
        process_disappeared_tracks()

        # Display object count
        print_object_count(annotator)

         # Send the annotated frame to the display/save queue
        try:
            display_queue.put(annotator.result())
            logging.debug("Processed frame added to display queue.")
        except mp.queues.Full:
            logging.warning("Display queue full, frame could not be added.")

    logging.info("Finished frame processing.")

def print_object_count(annotator=None):
    """Print the current object count."""
    logging.info("\nCurrent object count:")
    y_offset = 30
    for cls, count in object_count.items():
        class_name = model.names[cls] if model.names else f'cls {cls}'
        logging.info(f'{class_name}: {count}')

        if annotator:
            annotator.text((10, y_offset), f'{class_name}: {count}', txt_color=(255, 255, 255))
            y_offset += 30

def display_and_save(display_queue):
    """Function that handles displaying and saving processed frames"""
    logging.info("Starting frame display and save process...")
    global output_video_writer

    while not stop_event.is_set() or not display_queue.empty():
        try:
            frame = display_queue.get(timeout=0.1)  # Get frame from display queue
            logging.debug("Frame retrieved from the queue for display/save.")
        except mp.queues.Empty:
            logging.debug("Display queue empty, waiting for more frames.")
            continue

        if args.show:
            cv2.imshow('Tracking', frame)
            logging.debug("Displaying frame in window.")
            
            # Capture if 'q' is pressed
            if cv2.waitKey(1) & 0xFF == ord('q'):
                logging.info("Key 'q' pressed. Stopping display process.")
                stop_event.set()  # Stop everything if 'q' is pressed
                break

        if output_video_writer is not None:
            output_video_writer.write(frame)
            logging.debug("Frame saved to video file.")

    if output_video_writer is not None and output_video_writer.isOpened():
        output_video_writer.release()
        logging.info("Video file saved and closed successfully.")

    cv2.destroyAllWindows()
    logging.info("Display and save process finished.")

# Command-line arguments and model, database configuration, etc.
parser = argparse.ArgumentParser(description="Script for video processing and object detection with YOLO.")
parser.add_argument('video_source', type=str, help='Path to video file or RTSP URL')
parser.add_argument('--output', type=str, default=None, help='Path to output video file (optional)')
parser.add_argument('--model', type=str, default=None, help='Path to YOLO model file')
parser.add_argument("--mongo-uri", type=str, default=None, help="MongoDB URI")
parser.add_argument("--mongo-db-name", type=str, default="vista-count-yolov", help="MongoDB database name")
parser.add_argument("--clickhouse-host", type=str, default=None, help="ClickHouse connection HOST")
parser.add_argument("--clickhouse-port", type=str, default=None, help="ClickHouse connection PORT")
parser.add_argument("--cam-id", type=str, default=None, help="Camera ID (optional)")
parser.add_argument("--no-sync", action="store_true", help="Disable database synchronization")
parser.add_argument('--show', action='store_true', default=False, help='Display results in a window')

args = parser.parse_args()

video_source = args.video_source
output_video_path = args.output

model_path = args.model
model = YOLO(model_path)

if args.mongo_uri and not args.no_sync:
    mongo_client = MongoClient(args.mongo_uri)

if args.clickhouse_host and args.clickhouse_port and not args.no_sync:
    clickhouse_client = clickhouse_connect.get_client(host=args.clickhouse_host, port=args.clickhouse_port)

# If an output path is provided, initialize VideoWriter
if output_video_path:
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    output_video_writer = cv2.VideoWriter(output_video_path, fourcc, DEFAULT_FPS, (video_resolution["width"], video_resolution["height"]))

# Multiprocessing usage
if __name__ == '__main__':
    DEFAULT_SIZE_QUEUE = 750
    # Define multiprocessing queues
    frame_queue = mp.Queue(maxsize=DEFAULT_SIZE_QUEUE)
    display_queue = mp.Queue(maxsize=DEFAULT_SIZE_QUEUE)

    # Create processes
    capture_process = mp.Process(target=capture_frames, args=(video_source, frame_queue, DEFAULT_SIZE_QUEUE), name="Capture")
    processing_process = mp.Process(target=process_frames_sequentially, args=(frame_queue, display_queue, model), name="Processing")
    display_process = mp.Process(target=display_and_save, args=(display_queue,), name="Display")

    processes = [capture_process, processing_process, display_process]

    # Assign signal handler to stop processes
    signal.signal(signal.SIGINT, lambda sig, frame: signal_handler(sig, frame, processes))
    signal.signal(signal.SIGTERM, lambda sig, frame: signal_handler(sig, frame, processes))

    # Start processes
    for process in processes:
        process.start()

    # Wait for processes to finish
    for process in processes:
        process.join()

[glenn-jocher]

@tastafur it seems like your queue is growing due to the slower processing speed of the tracker. Consider reducing the frame rate of your video source or increasing the processing speed by optimizing your model or hardware. You might also explore using a more efficient tracker configuration. For further details, please refer to the Ultralytics tracking documentation.

[tastafur]

Hello @glenn-jocher
I have tried basing a nano and small model and the same thing keeps happening, quantified at fp16, I have not tried to lower the fps of the rstp transmission, I will try tomorrow, for example with the predictor it goes perfect but with the tracking I have been sticking a few days and this happens, I will also try to touch the tracking configuration of the yml that I have already been testing. The jetson we use is this reComputer Industrial J4012, Orin NX 16GB of 100 Tops and testing with the devkit of envy of urin I think the same thing happens, while it has memory the queue collects all the frames and goes perfect but as the reading of frames is faster than the traking fills the queue and I end up with no memory. I tried to simulate also to force it with sleep to try that the input simulate to go to the fps I wanted but I ended up losing fps. Could you give me some more ideas on how to solve this problem? Thanks, best regards

[tastafur]

@glenn-jocher I'm either Spanish or you live in Spain what a coincidence I'm Spanish haha

[glenn-jocher]

@tastafur thank you for reaching out! It's great to connect with fellow enthusiasts. If you have any questions about using Ultralytics YOLO for tracking, feel free to ask here. For more detailed guidance, please refer to our documentation.

[tastafur]

Good evening, I'm trying to configure the botsort file for when there is a drop of water in the bed that makes it lose the vehicle tracking and detect it again, I've been playing with the parameters for a while and I can't get it to associate the same trackId again, because of the damn drop of water, thank you very much for the help.
And it seems that the parameter does nothing,track_buffer I have set it up to 1000 and I have not seen any change.

# Ultralytics YOLO 🚀, AGPL-3.0 license
# Default YOLO tracker settings for BoT-SORT tracker https://github.com/NirAharon/BoT-SORT

tracker_type: botsort # tracker type, ['botsort', 'bytetrack']
track_high_thresh: 0.5 # threshold for the first association
track_low_thresh: 0.1 # threshold for the second association
new_track_thresh: 0.6 # threshold for init new track if the detection does not match any tracks
track_buffer: 100 # buffer to calculate the time when to remove tracks
match_thresh: 0.85 # threshold for matching tracks
fuse_score: True # Whether to fuse confidence scores with the iou distances before matching
# min_box_area: 10  # threshold for min box areas(for tracker evaluation, not used for now)

# BoT-SORT settings
gmc_method: sparseOptFlow # method of global motion compensation
# ReID model related thresh (not supported yet)
proximity_thresh: 0.5
appearance_thresh: 0.25
with_reid: True

[glenn-jocher]

@tastafur it seems like the track_buffer parameter isn't having the desired effect. You might want to experiment with track_high_thresh and new_track_thresh to improve track association. For more details, please refer to the tracking documentation.

[tastafur]

I was trying to upload, download all yesterday and nothing always came back to create a new trackId, today I will spend the day trying again, to see if I find the configuration, I ask, is it worth instead of using your implementation of the tracking take your detector and take the BoT-SORT library directly and implement it by hand, if I do not get it, this morning Spanish time, I will surely try by hand to see if I get better results.

[glenn-jocher]

If you're experiencing issues with track re-association due to water drops, adjusting track_high_thresh and new_track_thresh might help. Using the BoT-SORT library directly could offer more control, but ensure your configuration aligns with your specific needs. For further guidance, please refer to the tracking documentation.

[tastafur]

Good morning,
I think I found a bug but as I'm not so sure, I'll pass you the line. I use a jetson to track two lanes and for that I put the streaming at 15 fps and at a certain resolution so that everything goes well. But when trying to configure all the parameters of the BoT-SORT tracker there is one that no matter what I did apparently nothing happened and it was the track_buffer. Trying to make the tracker work as good as possible I found that in your implementation you have hardcoded that the tracker takes as reference 30 fps, so I think that's why the tracker and the track_buffer are not working well because it depends on the fps that the tracker has hardcoded and I can not pass the 15 fps xD

And here is the line where you have the fps harcode

https://github.com/ultralytics/ultralytics/blob/main/ultralytics/trackers/track.py#L45

[glenn-jocher]

@tastafur thank you for pointing this out. We'll review the hardcoded FPS reference. Meanwhile, you might consider adjusting the tracker's configuration to better suit your setup. For further details, please refer to our documentation on tracker configurations.

[guillepabon]

I've been running this code for a couple of weeks and everything was fine until the last update of the Ultralytics library. I have an object counter declared in the following way:

counter = solutions.ObjectCounter(
    view_img=False,
    reg_pts=region_points,
    names=model.names,
    draw_tracks=True,
    line_thickness=2,
)

Then, I define the type of tracker to be used with a custom model:

 tracks = model.track(
        frame,
        persist=True,
        conf=0.6,  # Confidence threshold
        tracker="bytetrack.yaml",
        show=False,
        classes=classes_to_count,
        verbose=False
    )

And finally, I am accessing every frame using the 'start_counting' function, specifying the frame to be analyzed and the track type:

frame = counter.start_counting(frame, tracks)

When I try to execute the code now, I get the error:

Traceback (most recent call last):
  File "/.../counting_video7_transformer.py", line 171, in <module>
    frame = counter.start_counting(frame, tracks)
AttributeError: 'ObjectCounter' object has no attribute 'start_counting'

I've attempted to update the method from 'start_counting' to 'count' (according to your limited documentation above). However, I get this error:

Traceback (most recent call last):
  File "/.../counting_video7_transformer.py", line 171, in <module>
    frame = counter.count(frame, tracks)
TypeError: ObjectCounter.count() takes 2 positional arguments but 3 were given

Any hints on how this should be handled moving forward? Thanks

[guillepabon]

Forgot to specify, I'm using custom trained YOLOv8, YOLOv10, and YOLOv11 models. And my previous code is not working anymore :S

[glenn-jocher]

@guillepabon it seems like the ObjectCounter class has changed. Please check the latest documentation or source code for the correct method signature. If count() is the new method, ensure you're passing the correct number of arguments. If issues persist, consider reviewing recent changes in the library that might affect your implementation.

[BlackiePiggy]

Hello, thank you very much for your work on computer vision.

I had a problem in development.

For the same test video, the same model is used. When I use predict mode and track mode, there is a large performance gap in target recognition. This is reflected in that fewer target boxes are detected in track mode than in predict mode.

I used the default byteTrack tracker or BoT Tracker. I have adjusted the tracker's hyperparameters for many times, but the problem has not been solved. May I ask why?

[BlackiePiggy]

I can provide my predict output at conf=0.5 and the track output(same conf=0.5) using the default bytetrack parameter. As down below, you can see, the ball detected were many times ignored in track mode.

predict output:

(base) evsjtu2@evsjtu2:~/disk2/ljy/obj_detect$ python predict.py
WARNING ⚠️ Python>=3.10 is required, but Python==3.8.5 is currently installed


WARNING ⚠️ inference results will accumulate in RAM unless `stream=True` is passed, causing potential out-of-memory
errors for large sources or long-running streams and videos. See https://docs.ultralytics.com/modes/predict/ for help.

Example:
    results = model(source=..., stream=True)  # generator of Results objects
    for r in results:
        boxes = r.boxes  # Boxes object for bbox outputs
        masks = r.masks  # Masks object for segment masks outputs
        probs = r.probs  # Class probabilities for classification outputs

video 1/1 (frame 1/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 29.8ms
video 1/1 (frame 2/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 3/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 4/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 5/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.8ms
video 1/1 (frame 6/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 7/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 8/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 9/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 19.6ms
video 1/1 (frame 10/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 11/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 12/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 19.6ms
video 1/1 (frame 13/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.2ms
video 1/1 (frame 14/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 15/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 16/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 17/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 18/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 19/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.3ms
video 1/1 (frame 20/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 21/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 22/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 23/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 24/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 25/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 26/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 27/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 28/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 17.2ms
video 1/1 (frame 29/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 30/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 31/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 32/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 18.4ms
video 1/1 (frame 33/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 34/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 17.7ms
video 1/1 (frame 35/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.3ms
video 1/1 (frame 36/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.4ms
video 1/1 (frame 37/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.6ms
video 1/1 (frame 38/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.9ms
video 1/1 (frame 39/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.6ms
video 1/1 (frame 40/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 41/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.6ms
video 1/1 (frame 42/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.4ms
video 1/1 (frame 43/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.6ms
video 1/1 (frame 44/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.4ms
video 1/1 (frame 45/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.6ms
video 1/1 (frame 46/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.6ms
video 1/1 (frame 47/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.5ms
video 1/1 (frame 48/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.7ms
video 1/1 (frame 49/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.5ms
video 1/1 (frame 50/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.5ms
video 1/1 (frame 51/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.9ms
video 1/1 (frame 52/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 16.0ms
video 1/1 (frame 53/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.7ms
video 1/1 (frame 54/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.9ms
video 1/1 (frame 55/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 56/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 57/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 58/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.4ms
video 1/1 (frame 59/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.6ms
video 1/1 (frame 60/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.4ms
video 1/1 (frame 61/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.8ms
video 1/1 (frame 62/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.6ms
video 1/1 (frame 63/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.6ms
video 1/1 (frame 64/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.4ms
video 1/1 (frame 65/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.4ms
video 1/1 (frame 66/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 2 rims, 15.3ms
video 1/1 (frame 67/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 2 humans, 15.7ms
video 1/1 (frame 68/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 2 humans, 15.4ms
video 1/1 (frame 69/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 2 humans, 15.5ms
video 1/1 (frame 70/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 2 humans, 15.3ms
video 1/1 (frame 71/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 72/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 73/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 74/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 75/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 76/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 77/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 78/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.3ms
video 1/1 (frame 79/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 80/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.9ms
video 1/1 (frame 81/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 2 rims, 15.8ms
video 1/1 (frame 82/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 83/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 84/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 85/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 86/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 87/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 88/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 89/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 90/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 91/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 92/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 93/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 94/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 18.5ms
video 1/1 (frame 95/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.7ms
video 1/1 (frame 96/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.1ms
video 1/1 (frame 97/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 98/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 99/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 100/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 101/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 102/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 103/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 104/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 105/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 106/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 107/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 108/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.5ms
video 1/1 (frame 109/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 110/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 111/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 112/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 113/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 114/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 19.1ms
video 1/1 (frame 115/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 116/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 117/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.2ms
video 1/1 (frame 118/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.4ms
video 1/1 (frame 119/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 120/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.4ms
video 1/1 (frame 121/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.4ms
video 1/1 (frame 122/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 123/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 124/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 125/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 126/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 127/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 128/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 129/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 130/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 131/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 132/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 20.4ms
video 1/1 (frame 133/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.4ms
video 1/1 (frame 134/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 135/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 136/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 137/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 138/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 139/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 140/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 15.4ms
video 1/1 (frame 141/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 15.4ms
video 1/1 (frame 142/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 15.6ms
video 1/1 (frame 143/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 144/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 145/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 146/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 147/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 148/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 149/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 150/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 151/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.2ms
video 1/1 (frame 152/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 153/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 154/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.1ms
video 1/1 (frame 155/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 156/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 157/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.1ms
video 1/1 (frame 158/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 159/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 160/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 161/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 162/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 163/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 164/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 165/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 166/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 167/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.6ms
video 1/1 (frame 168/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 169/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 170/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 171/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 172/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.1ms
video 1/1 (frame 173/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.8ms
video 1/1 (frame 174/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 175/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.6ms
video 1/1 (frame 176/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 177/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.5ms
video 1/1 (frame 178/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 179/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 180/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 181/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 182/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 183/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 184/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 185/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 186/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 187/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 188/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.1ms
video 1/1 (frame 189/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 190/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 191/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 192/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 18.1ms
video 1/1 (frame 193/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 194/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 195/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 196/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 197/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 198/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 199/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 200/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 201/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 202/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 203/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 204/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.9ms
video 1/1 (frame 205/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 206/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 207/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 20.7ms
video 1/1 (frame 208/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 209/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 210/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 211/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 212/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 213/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 214/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 215/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.3ms
video 1/1 (frame 216/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.4ms
video 1/1 (frame 217/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 218/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 219/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 18.0ms
video 1/1 (frame 220/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 221/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 222/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 223/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 224/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 225/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 226/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 227/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 228/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.6ms
video 1/1 (frame 229/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 230/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 231/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 232/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 233/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 234/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.1ms
video 1/1 (frame 235/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 236/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 237/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 238/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 239/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.3ms
video 1/1 (frame 240/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 241/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 242/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 243/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 244/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 245/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 246/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.7ms
video 1/1 (frame 247/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.8ms
video 1/1 (frame 248/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 16.1ms
video 1/1 (frame 249/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 250/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.1ms
video 1/1 (frame 251/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 252/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 253/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 254/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 255/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 256/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 257/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 258/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 259/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 260/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 261/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 262/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 263/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 264/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 265/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.8ms
video 1/1 (frame 266/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 267/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 268/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.0ms
video 1/1 (frame 269/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 270/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 271/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 272/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 273/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 274/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 275/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 276/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 277/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 278/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 279/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 280/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 23.1ms
video 1/1 (frame 281/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.5ms
video 1/1 (frame 282/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.8ms
video 1/1 (frame 283/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 16.1ms
video 1/1 (frame 284/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.7ms
video 1/1 (frame 285/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.8ms
video 1/1 (frame 286/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 16.2ms
video 1/1 (frame 287/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 16.0ms
video 1/1 (frame 288/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.7ms
video 1/1 (frame 289/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 290/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 291/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 292/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.5ms
video 1/1 (frame 293/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.7ms
video 1/1 (frame 294/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 17.0ms
video 1/1 (frame 295/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 296/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.6ms
video 1/1 (frame 297/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.5ms
video 1/1 (frame 298/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.6ms
video 1/1 (frame 299/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.8ms
video 1/1 (frame 300/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 301/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 302/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 303/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 304/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 305/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 20.4ms
video 1/1 (frame 306/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 16.4ms
video 1/1 (frame 307/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 15.6ms
video 1/1 (frame 308/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 15.7ms
video 1/1 (frame 309/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 2 rims, 18.2ms
video 1/1 (frame 310/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 2 rims, 15.9ms
video 1/1 (frame 311/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 2 rims, 15.7ms
video 1/1 (frame 312/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 2 rims, 15.8ms
video 1/1 (frame 313/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 2 rims, 15.5ms
video 1/1 (frame 314/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 2 rims, 15.6ms
video 1/1 (frame 315/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 316/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 317/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 318/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 2 rims, 15.5ms
video 1/1 (frame 319/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 320/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 321/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 2 rims, 15.8ms
video 1/1 (frame 322/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 2 rims, 15.7ms
video 1/1 (frame 323/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.2ms
video 1/1 (frame 324/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 325/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 326/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 327/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 328/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 329/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 330/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 331/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 332/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 333/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.4ms
video 1/1 (frame 334/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 335/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 336/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 337/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 338/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 15.7ms
video 1/1 (frame 339/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 340/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 341/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 342/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 343/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 2 humans, 1 rim, 16.3ms
video 1/1 (frame 344/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.5ms
video 1/1 (frame 345/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 346/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 347/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 348/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 349/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 350/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 351/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 352/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 353/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 354/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 355/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 356/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 357/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 2 balls, 1 human, 1 rim, 15.5ms
video 1/1 (frame 358/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 359/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 360/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 361/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 362/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 363/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 364/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 2 rims, 15.8ms
video 1/1 (frame 365/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 366/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 367/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.2ms
video 1/1 (frame 368/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 369/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 370/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 371/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 372/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 373/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 374/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 375/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 376/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 377/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 378/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 379/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 380/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 381/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 382/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 383/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 17.4ms
video 1/1 (frame 384/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 385/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 386/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 387/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 388/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 389/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 390/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 391/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 392/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 393/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 17.4ms
video 1/1 (frame 394/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 16.4ms
video 1/1 (frame 395/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 15.5ms
video 1/1 (frame 396/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 2 rims, 15.8ms
video 1/1 (frame 397/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 17.6ms
video 1/1 (frame 398/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 399/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 400/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 401/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 402/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 403/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 404/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 405/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 406/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 407/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 408/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 409/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
Speed: 1.4ms preprocess, 15.9ms inference, 1.7ms postprocess per image at shape (1, 3, 640, 384)
Results saved to runs/detect/predict6

track output:

(base) evsjtu2@evsjtu2:~/disk2/ljy/obj_detect$ python track.py
WARNING ⚠️ Python>=3.10 is required, but Python==3.8.5 is currently installed


WARNING ⚠️ inference results will accumulate in RAM unless `stream=True` is passed, causing potential out-of-memory
errors for large sources or long-running streams and videos. See https://docs.ultralytics.com/modes/predict/ for help.

Example:
    results = model(source=..., stream=True)  # generator of Results objects
    for r in results:
        boxes = r.boxes  # Boxes object for bbox outputs
        masks = r.masks  # Masks object for segment masks outputs
        probs = r.probs  # Class probabilities for classification outputs

video 1/1 (frame 1/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 30.9ms
video 1/1 (frame 2/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 20.5ms
video 1/1 (frame 3/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 4/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 5/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 6/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 7/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 8/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 9/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 10/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 11/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 12/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 13/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 14/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 15/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 19.9ms
video 1/1 (frame 16/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.2ms
video 1/1 (frame 17/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 18/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 19/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.2ms
video 1/1 (frame 20/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.3ms
video 1/1 (frame 21/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 22/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 23/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 24/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 25/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 26/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 27/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 28/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 29/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 30/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 31/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 32/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 33/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 34/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 35/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 36/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 37/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 17.3ms
video 1/1 (frame 38/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.2ms
video 1/1 (frame 39/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 40/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 41/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 42/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.3ms
video 1/1 (frame 43/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 44/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 45/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 46/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 47/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 48/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 49/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 50/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.2ms
video 1/1 (frame 51/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.4ms
video 1/1 (frame 52/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 53/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 54/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.6ms
video 1/1 (frame 55/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 56/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 57/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.3ms
video 1/1 (frame 58/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 59/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 60/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 61/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 62/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 63/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 64/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 65/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 66/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 67/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 15.5ms
video 1/1 (frame 68/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 15.5ms
video 1/1 (frame 69/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 16.0ms
video 1/1 (frame 70/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 15.4ms
video 1/1 (frame 71/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 72/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 73/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 74/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 75/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 76/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 77/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 78/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 79/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 80/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 81/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 82/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 83/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 84/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 85/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 86/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 87/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 88/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 89/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 90/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 91/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 92/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 93/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 94/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 95/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 20.7ms
video 1/1 (frame 96/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.3ms
video 1/1 (frame 97/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 98/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 99/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 100/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 101/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 102/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 103/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 104/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 105/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 106/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 107/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.3ms
video 1/1 (frame 108/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.8ms
video 1/1 (frame 109/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 110/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 111/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 112/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 113/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 114/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 115/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 116/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 117/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 118/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 119/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.6ms
video 1/1 (frame 120/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 121/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 122/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 123/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 124/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 125/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 126/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.7ms
video 1/1 (frame 127/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 128/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 129/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 130/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 131/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 132/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 133/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 17.6ms
video 1/1 (frame 134/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 23.1ms
video 1/1 (frame 135/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 136/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 137/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 17.1ms
video 1/1 (frame 138/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 17.8ms
video 1/1 (frame 139/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.2ms
video 1/1 (frame 140/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 17.0ms
video 1/1 (frame 141/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 16.1ms
video 1/1 (frame 142/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 16.1ms
video 1/1 (frame 143/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.1ms
video 1/1 (frame 144/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 145/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 146/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 147/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 148/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.9ms
video 1/1 (frame 149/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.5ms
video 1/1 (frame 150/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 151/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 152/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.5ms
video 1/1 (frame 153/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 20.2ms
video 1/1 (frame 154/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 155/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 156/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 157/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 158/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 159/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.4ms
video 1/1 (frame 160/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 161/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 162/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.3ms
video 1/1 (frame 163/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 164/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.3ms
video 1/1 (frame 165/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.3ms
video 1/1 (frame 166/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 167/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 168/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 169/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 170/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 171/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 172/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.3ms
video 1/1 (frame 173/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 174/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 175/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 176/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 177/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 178/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 179/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.3ms
video 1/1 (frame 180/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 181/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 182/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 183/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 184/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 185/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 186/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 187/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 188/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 189/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 190/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 191/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 192/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.0ms
video 1/1 (frame 193/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 16.2ms
video 1/1 (frame 194/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 195/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 196/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.9ms
video 1/1 (frame 197/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 198/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.8ms
video 1/1 (frame 199/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 200/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 201/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 202/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 203/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 204/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 205/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 206/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 207/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 208/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 209/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 210/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 19.5ms
video 1/1 (frame 211/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 212/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 213/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 214/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 215/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 216/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 217/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 18.1ms
video 1/1 (frame 218/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.4ms
video 1/1 (frame 219/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 220/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 221/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 222/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 223/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 224/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 19.3ms
video 1/1 (frame 225/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 226/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 227/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 228/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 229/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 230/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 231/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 232/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 233/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 234/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 235/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.0ms
video 1/1 (frame 236/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 237/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 238/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 239/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 240/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 241/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 242/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 243/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 244/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 245/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 246/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 247/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 248/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 249/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 250/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 251/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 252/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 253/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 254/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 255/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 17.3ms
video 1/1 (frame 256/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 18.2ms
video 1/1 (frame 257/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 258/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 259/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 260/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 261/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 262/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 263/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 264/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 265/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 266/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 267/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 268/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 269/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 270/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.3ms
video 1/1 (frame 271/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 272/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 273/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 274/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 275/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 276/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 277/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 278/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 279/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.0ms
video 1/1 (frame 280/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 281/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 282/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 283/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.8ms
video 1/1 (frame 284/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 285/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 286/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 287/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 288/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 289/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.0ms
video 1/1 (frame 290/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 291/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 292/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 293/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 294/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.7ms
video 1/1 (frame 295/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.3ms
video 1/1 (frame 296/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 297/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.1ms
video 1/1 (frame 298/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 299/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 300/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 301/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.5ms
video 1/1 (frame 302/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.6ms
video 1/1 (frame 303/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 304/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 15.5ms
video 1/1 (frame 305/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 15.8ms
video 1/1 (frame 306/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 16.3ms
video 1/1 (frame 307/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 16.3ms
video 1/1 (frame 308/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 15.8ms
video 1/1 (frame 309/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 310/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 311/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.2ms
video 1/1 (frame 312/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 313/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 314/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 315/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 316/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 317/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 318/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 319/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 320/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 321/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 18.1ms
video 1/1 (frame 322/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 323/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 324/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 325/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 326/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 327/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 328/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 329/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 330/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 331/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 332/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.5ms
video 1/1 (frame 333/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.4ms
video 1/1 (frame 334/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 335/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.6ms
video 1/1 (frame 336/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 337/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 19.5ms
video 1/1 (frame 338/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 339/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 340/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 16.0ms
video 1/1 (frame 341/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 rim, 15.6ms
video 1/1 (frame 342/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 343/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 344/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 345/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 346/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.5ms
video 1/1 (frame 347/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.4ms
video 1/1 (frame 348/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.6ms
video 1/1 (frame 349/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 15.7ms
video 1/1 (frame 350/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 21.3ms
video 1/1 (frame 351/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 19.4ms
video 1/1 (frame 352/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 20.3ms
video 1/1 (frame 353/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 18.6ms
video 1/1 (frame 354/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 ball, 1 human, 1 rim, 18.5ms
video 1/1 (frame 355/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.2ms
video 1/1 (frame 356/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.4ms
video 1/1 (frame 357/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 358/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 359/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 360/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.9ms
video 1/1 (frame 361/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 362/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 363/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 364/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.6ms
video 1/1 (frame 365/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 366/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 367/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 20.9ms
video 1/1 (frame 368/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 17.0ms
video 1/1 (frame 369/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 370/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 371/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.9ms
video 1/1 (frame 372/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.0ms
video 1/1 (frame 373/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 374/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 375/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 376/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 17.5ms
video 1/1 (frame 377/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.8ms
video 1/1 (frame 378/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 379/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 380/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 381/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.7ms
video 1/1 (frame 382/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 383/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.2ms
video 1/1 (frame 384/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.6ms
video 1/1 (frame 385/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 (no detections), 16.2ms
video 1/1 (frame 386/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.4ms
video 1/1 (frame 387/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 20.9ms
video 1/1 (frame 388/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.0ms
video 1/1 (frame 389/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.7ms
video 1/1 (frame 390/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.8ms
video 1/1 (frame 391/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 392/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.0ms
video 1/1 (frame 393/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.0ms
video 1/1 (frame 394/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 (no detections), 16.0ms
video 1/1 (frame 395/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 (no detections), 16.1ms
video 1/1 (frame 396/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 (no detections), 15.7ms
video 1/1 (frame 397/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.9ms
video 1/1 (frame 398/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.8ms
video 1/1 (frame 399/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 20.3ms
video 1/1 (frame 400/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 401/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 402/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 15.9ms
video 1/1 (frame 403/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.3ms
video 1/1 (frame 404/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.4ms
video 1/1 (frame 405/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 human, 1 rim, 16.1ms
video 1/1 (frame 406/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.3ms
video 1/1 (frame 407/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 16.4ms
video 1/1 (frame 408/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.4ms
video 1/1 (frame 409/409) /home/evsjtu2/disk2/ljy/obj_detect/testdata/test.mp4: 640x384 1 rim, 15.2ms
Speed: 1.5ms preprocess, 16.0ms inference, 1.7ms postprocess per image at shape (1, 3, 640, 384)
Results saved to runs/detect/track23

[glenn-jocher]

@BlackiePiggy it seems the issue might be related to the tracker's confidence threshold settings. In tracking mode, objects with confidence scores below track_high_thresh may not be tracked. Try adjusting this parameter in your tracker configuration file. For more details, please refer to the Ultralytics tracking documentation.

[clearpaint]

I'm trying to use the above example to track hockey players.
Note I did use Roboflow/Yolo to train my model , it can be found here:
https://universe.roboflow.com/finalproject-l7mc2/hockey-jerseys/

I have tried many different configurations without great success. (ByteTrack and Botsort) As well as trying to use norrfair.
Notably, between frames the crossing of players sometimes (often) the ID will swap from one player to another.
Also , when a player goes briefly out of view and comes back into view there is a new ID assigned.

If you watch the video you can see ID:4 changes many times after occlusion with another player.
And again this happens when the players goes into the corner just out of view.
I'm trying to understand what setting knobs can be used to help with this.

I have uploaded my examples to https://github.com/clearpaint/yolo/
Any help or suggestion would be greatly appreciated.

[glenn-jocher]

@clearpaint to address the ID switching and reassignment issues, consider adjusting the track_high_thresh parameter in your tracker configuration file to improve ID consistency. Additionally, experimenting with different tracker algorithms like BoT-SORT or ByteTrack and fine-tuning their parameters may help. For further customization, you can modify the tracker configuration files available in the Ultralytics repository.

[clearpaint]

so tired some different setting # Ultralytics YOLO 🚀, AGPL-3.0 license

Optimized YOLO tracker settings for improved ID consistency with ByteTrack tracker

https://github.com/ifzhang/ByteTrack

tracker_type: bytetrack # tracker type, ['botsort', 'bytetrack']

Tracking thresholds

track_high_thresh: 0.5 # Increased threshold for the first association
track_low_thresh: 0.3 # Increased threshold for the second association
new_track_thresh: 0.5 # Increased threshold for initiating new tracks

track_buffer: 60 # Extended buffer to maintain tracks longer
match_thresh: 0.6 # Decreased threshold for matching tracks

fuse_score: False # Disabled score fusion to prevent misassociation
min_box_area: 20 # Set minimum box area to filter out small detections

and :

Ultralytics YOLO 🚀, AGPL-3.0 license

Optimized YOLO tracker settings for improved ID consistency with BoT-SORT tracker

https://github.com/NirAharon/BoT-SORT

tracker_type: botsort # tracker type, ['botsort', 'bytetrack']

Tracking thresholds

track_high_thresh: 0.4 # Increased threshold for the first association
track_low_thresh: 0.2 # Increased threshold for the second association
new_track_thresh: 0.4 # Increased threshold for initiating new tracks

track_buffer: 60 # Extended buffer to maintain tracks longer
match_thresh: 0.6 # Decreased threshold for matching tracks

fuse_score: False # Disabled score fusion to prevent misassociation
min_box_area: 20 # Set minimum box area to filter out small detections

BoT-SORT settings

gmc_method: sparseOptFlow # Method of global motion compensation

ReID model settings

proximity_thresh: 0.4 # Adjusted proximity threshold for spatial matching
appearance_thresh: 0.3 # Adjusted appearance threshold for ReID matching
with_reid: True # Enabled ReID module for better ID consistency

neither yielded much better results.

[glenn-jocher]

It sounds like you're facing challenges with ID consistency in your tracking setup. Consider adjusting the track_high_thresh and track_buffer parameters to improve ID retention. Additionally, enabling the ReID module in BoT-SORT might help maintain consistent IDs across occlusions. For further tuning, you might explore different combinations of proximity_thresh and appearance_thresh. If you have any specific questions or need further assistance, feel free to ask!

[clearpaint]

Hi @glenn-jocher , Is it possible for someone to help try run this example? I have tried many configuration all without great success. Most of the changes have actually made it worse. Its curious how good the example looks maybe its just the speed and repeated occlusions are just not trackable? https://www.youtube.com/watch?v=OrTciwKdSdA https://www.youtube.com/embed/OrTciwKdSdA?si=QrNWbCytZV3ebWka

[glenn-jocher]

It seems like you're encountering challenges with maintaining consistent tracking IDs. This could be due to rapid movements or occlusions in your video. Consider fine-tuning parameters like track_high_thresh or match_thresh and enabling ReID to improve ID consistency. You can also explore different trackers like BoT-SORT or ByteTrack with custom configurations to see if they offer better results. If you'd like more information on specific configurations, check out the Ultralytics YOLO Track Mode documentation.

[tofulim]

Hi
we got some problem with tracked object's id
when tracking run for a long time object id would be huge.

we found that object id fluctuating within 19000 - 24000
object's id changes several times in a few sec
i think that obj id has overflowed

is there any solution or controllable variable?

[tofulim]

Solved by link

model.predictor.trackers[0].reset()

[glenn-jocher]

@tofulim thank you for sharing your observations. The issue with fluctuating object IDs could be related to the tracker configuration or the confidence threshold settings. You might want to adjust the track_high_thresh parameter in your tracker configuration file to see if it stabilizes the IDs. For more detailed guidance, please refer to the Ultralytics tracking documentation.

[clearpaint]

@tofulim , I don't understand what you mean ? How does using reset correct the ids?

[tofulim]

hi @clearpaint , i didn't mean that using reset correct object's id. but it prevent object id's inconsistency

my hypothesis was when object_id getting bigger, it overflowed. at first i tested with short mp4 files, infinitely streaming with RTSP. i found that object_id was consistent but after some times passed(more than 1 day) object id increased over 18k and i found that object_id's fluctuation.
for example, my video has a scene of a guy passing the street. so the object_id should be consistent till he complete passing the street and get out of the screen. but as i found, guy's object_id changed dynamic

so i reset the tracker when object_id reached over 10k. then obejct_id changing has gone. (i tested this morning)

[tofulim]

sry, our engine has fast FPS(nearby 7) when it initialized then slowly converge to lower FPS(nearby 5). and i found that on lower FPS we got track id fluctuation because each frames of sequence has more time gap. so i thought increasing track_id was the problem but it was not true. time gap on input frame sequence was the problem.

so glenn was right

@NasserAlqeifi it sounds like the tracker may be losing track of the object between frames, which can cause inconsistent IDs. You can try adjusting the track_high_thresh parameter in your tracker configuration to improve object ID consistency. Additionally, using a more robust tracker like ByteTrack might help. If issues persist, consider exploring the Ultralytics tracker configuration options for further customization.

[NasserAlqeifi]

I have a problem with the tracker. I trained Segmentation Model 11, and tried the tracker with the model, but the tracker gives different IDs for the same object. How can I fix that? Are there any tips to solve the problem?

Note: I read all the documents on the site but did not find a solution to my problem

thank you for your help !

[glenn-jocher]

@NasserAlqeifi it sounds like the tracker may be losing track of the object between frames, which can cause inconsistent IDs. You can try adjusting the track_high_thresh parameter in your tracker configuration to improve object ID consistency. Additionally, using a more robust tracker like ByteTrack might help. If issues persist, consider exploring the Ultralytics tracker configuration options for further customization.

[guetaro]

Hi,

I'm using yolo11, with weights I trained myself.

For some reason, the predict and track results are completely the same:

model = YOLO(weights)
while true:
ret, frame = cap.read()
if not ret:
break
frame_test = frame.copy()
output = model.track(source =frame, conf = conf, imgsz = imgsz , persist = True, tracker = 'bytetrack.yaml')
for det in range (output [0].boxes.xyxy):
**** plotting ****
output_test = model.predict(source = frame_test, conf = conf, imgsz = imgsz)
for det in range (output_test[0].boxes.xyxy):
**** plotting ****

After printing / plotting the annotations, it seems it exactly the same.
The bytetracker should have filtered many FP annotations, but it doesn't.
I tried to run only one process at the time (command the track / predict and run the other only), and it still the same.
Moreover, tried to create two model instances - still the same.
Any ideas why it doesn't work?

Thank you in advance!

[glenn-jocher]

@guetaro it sounds like the ByteTrack tracker might not be configured or functioning as expected. Please check the bytetrack.yaml file for any default settings that might need adjustments, such as confidence thresholds or IOU parameters, which could affect tracking behavior. Additionally, ensure your model's weights are compatible with the tracking setup. For further insights, you may want to explore using a different tracker like BoT-SORT to compare results. For more details on configuration, refer to the Ultralytics Track documentation.

[clearpaint]

@glenn-jocher you keep referencing the "Documentation" but its not really documented how each parameter is to be used or interpreted. The answer seems to always be "play with the parameters."

[glenn-jocher]

@clearpaint thank you for your feedback. We understand your concern, and while experimenting with parameters is often necessary due to the variability of use cases, we're continuously working to enhance our documentation. For specific guidance, please visit the Track Mode documentation for examples and detailed parameter descriptions.

[guetaro]

Hi @glenn-jocher,

Thanks for your answer.

I kept debugging this issue, and maybe got my solution.

It seems the tracker's output contain both tracked and predicted results (results from predict and from tracker),
but only if results[0].boxes.is_track == True it means it is a valid track and not just regular prediction.

Am I right?

Thanks again.

[glenn-jocher]

Yes, you're correct. In the tracking results, results[0].boxes.is_track == True indicates that the detection is a valid track rather than just a standalone prediction. This differentiation helps in filtering out predictions that are not part of a tracked sequence. For further details on tracking configuration, consider reviewing the Track documentation.

[tylerfrisinger01]

I have a model trained on about 1200 images (including augmented images) my project is to detect if there are 0,1 or 2 monkeys within each frame of the video. However if for example there is only one monkey seen in the frame the label will switch randomly between the two label names I have (for now we will say “monkey1” and “monkey2”). This is a problem for me because I save all the cropped images of each label and play it as a video for a closeup view. So then my cropped images will not be consistent. I have done the model.track(persist=True, agnostic_nms=True, iou=0.50) in order to try and reduce the likelihood of this happening however it still does more often then I would like. What options do I have in order to fix this or reduce the likelihood? I am also using Yolov11 fyi. Thank you so much for anything you come up with I’ve been stuck for weeks

[glenn-jocher]

@tylerfrisinger01 it sounds like you're encountering issues with label consistency during tracking. To improve this, you might try using a more robust tracker like ByteTrack, which can be enabled with tracker="bytetrack.yaml", as it may handle object ID consistency better. Additionally, ensure your training data is balanced and includes varied examples for each class. You can also experiment with adjusting the tracking parameters or refining your model to enhance performance. For more configuration options, visit the Ultralytics Track documentation.

[Ranjith3436]

i want to know the yolo model can process every resolution images as it is. or the process only for decided resolution (640,640). do you convert image size before the inference internally?

[glenn-jocher]

@Ranjith3436 ultralytics YOLO models internally resize input images to a default resolution (e.g., 640x640 for most configurations) before inference, ensuring consistency in preprocessing and performance optimization. This resizing happens automatically, so users can input images of any resolution without manual adjustments.

[ds-dinis007]

Can we use tracker mode other than bot sort and bytetrack? for example with deepsort tracker.
how to configure it?

[glenn-jocher]

@ds-dinis007 currently, Ultralytics YOLO supports BoT-SORT and ByteTrack trackers out of the box. To use a custom tracker like DeepSORT, you would need to modify or create a new tracker configuration file in the ultralytics/cfg/trackers directory and implement the necessary integration. Please refer to the contributing guide if you'd like to add support for a new tracker to the repository.