novel deep learning research works with PaddlePaddle

[ECCV'2020] STTN: Learning Joint Spatial-Temporal Transformations for Video Inpainting

Formation Flight in Dense Environments

STEP: Spatio-Temporal Progressive Learning for Video Action Detection. CVPR'19 (Oral)

State-of-the-art model for person re-identification in Multi-camera Multi-Target Tracking. Benchmarked on Market-1501 and DukeMTMTC-reID datasets.

FedSTIL: Spatial-Temporal Federated Learning for Lifelong Person Re-identification on Distributed Edges. (TCSVT'23)

[EMNLP 2024] A Video Chat Agent with Temporal Prior

A high-level Python framework to evaluate the skill of geospatial datasets by comparing candidates to benchmark maps producing agreement maps and metrics.

A Python framework to evaluate geospatial datasets by comparing candidate and benchmark maps to compute agreement maps and statistics.

MATLAB implementation of ST-Matching Algorithm for map matching

Paper: Task-Adaptive Meta-Learning Framework for Advancing Spatial Generalizability.

A unified Spatial Temporal Net Trainer for different baselines in Spatial Temporal Forecasting problems.

A framework for designing and evaluating Spatio-Temporal networks.

Pytorch version of ST-ResNet

This repository contains the source code for our Senior Capstone Project, a cornerstone of the Data Science major at UC San Diego.

Decomposing Time-Lapse Paintings into Layers

The LC2019 team winning entry for the EVA 2019 data competition

PredRNN implementation using Tensorflow.

A Python implementation of sti-dbscan: an algorithm based on st-dbscan (which is based on dbscan) that clusters spatial-temporal interval data

Some guides using postgis to load and processing environment-related data