Example using mmyolo and yolo-pose

[escorciav]

Is there an example using YOLO-POSE in mmpose?
If not, what about any recent YOLO backbone.

Could you please discuss if that will happen in the future?
Any tip or advice on how to do so is highly appreciated.

Thanks a lot for providing & maintaining this awesome codebase, model zoo, and benchmark.

[jin-s13]

Due to license issues, we will not support YOLO-related algorithms.
Maybe mmyolo could support it.

[jin-s13]

further discussion in open-mmlab/mmyolo#136

[Bovey0809]

I made one yolox-pose based on mmyolo. https://github.com/Bovey0809/mmyolo-pose

[nanhui69]

现在可以支持yolov5、yolov8么？我这报错，KeyError: 'YOLOv5KeepRatioResize is not in the transform registry. Please check whether the value of YOLOv5KeepRatioResize is correct or it was registered as expected. More details can be found at https://mmengine.readthedocs.io/en/latest/advanced_tutorials/config.html#import-the-custom-module ，在mmpose里调用inference_detector(detector, img) 出现错误，奇怪的是我在其它目录调用init_detector 不会出错_可以推理，怎么解决

[Ben-Louis]

现在可以支持yolov5、yolov8么？我这报错，KeyError: 'YOLOv5KeepRatioResize is not in the transform registry. Please check whether the value of YOLOv5KeepRatioResize is correct or it was registered as expected. More details can be found at https://mmengine.readthedocs.io/en/latest/advanced_tutorials/config.html#import-the-custom-module ，在mmpose里调用inference_detector(detector, img) 出现错误，奇怪的是我在其它目录调用init_detector 不会出错_可以推理，怎么解决

可以试下 inferencer:

python demo/inferencer_demo.py tests/data/coco/000000197388.jpg --pose2d human \
--det-model ../../repos/mmyolo/configs/yolov5/yolov5_m-v61_syncbn_fast_8xb16-300e_coco.py \
--det-weights https://download.openmmlab.com/mmyolo/v0/yolov5/yolov5_m-v61_syncbn_fast_8xb16-300e_coco/yolov5_m-v61_syncbn_fast_8xb16-300e_coco_20220917_204944-516a710f.pth \
--vis-out-dir vis_results/yolo

[nanhui69]

现在可以支持yolov5、yolov8么？我这报错，KeyError: 'YOLOv5KeepRatioResize is not in the transform registry. Please check whether the value of YOLOv5KeepRatioResize is correct or it was registered as expected. More details can be found at https://mmengine.readthedocs.io/en/latest/advanced_tutorials/config.html#import-the-custom-module ，在mmpose里调用inference_detector(detector, img) 出现错误，奇怪的是我在其它目录调用init_detector 不会出错_可以推理，怎么解决

可以试下 inferencer:

python demo/inferencer_demo.py tests/data/coco/000000197388.jpg --pose2d human \
--det-model ../../repos/mmyolo/configs/yolov5/yolov5_m-v61_syncbn_fast_8xb16-300e_coco.py \
--det-weights https://download.openmmlab.com/mmyolo/v0/yolov5/yolov5_m-v61_syncbn_fast_8xb16-300e_coco/yolov5_m-v61_syncbn_fast_8xb16-300e_coco_20220917_204944-516a710f.pth \
--vis-out-dir vis_results/yolo

这个是测试yolov5还是？我这边目前是用的yolov8

[Ben-Louis]

这个是测试yolov5还是？我这边目前是用的yolov8

上面这个命令是调用 yolov5 的例子，你可以换成 yolov8 的 config 和 checkpoint
更多细节可以参考 Inferencer 文档

[nanhui69]

这个是测试yolov5还是？我这边目前是用的yolov8

上面这个命令是调用 yolov5 的例子，你可以换成 yolov8 的 config 和 checkpoint 更多细节可以参考 Inferencer 文档

mmpose 是master分支么，还是我现在的dev-1.x 也支持呢？

[nanhui69]

这个是测试yolov5还是？我这边目前是用的yolov8

上面这个命令是调用 yolov5 的例子，你可以换成 yolov8 的 config 和 checkpoint 更多细节可以参考 Inferencer 文档

请问这个检测器有多个类别的话，在哪进行过滤呢（只选择行人）