Jinkun Cao's repositories
Pixel2Mesh
A complete Pixel2Mesh implementation in PyTorch
Detailed-VideoAvatar
Reproduced Detailed-VideoAvatar project for 3D Human-Body Reconstruction
disentanglement_lib_med
[NeurIPS 2022] disentanglement evaluation robust to model dimension variance.
mmtracking
OpenMMLab Video Perception Toolbox. It supports Video Object Detection (VID), Multiple Object Tracking (MOT), Single Object Tracking (SOT), Video Instance Segmentation (VIS) with a unified framework.
SJTU-SE-assignments
Repo for the assignments from the Software Engineering major of Shanghai Jiao Tong University
DanceTrack
DanceTrack: Multiple Object Tracking in Uniform Appearance and Diverse Motion
mmsegmentation
OpenMMLab Semantic Segmentation Toolbox and Benchmark.
slow_disentanglement
Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding
SurviveSJTUManual
更新2008年版本的《上海交通大学生存手册》gitbook发布于https://survivesjtu.gitbook.io/survivesjtumanual/
TransTrack
Multiple-Object Tracking with Transformer
Voice-Chinese-Poker
欢乐语音斗地主!
humor
Code for ICCV 2021 paper "HuMoR: 3D Human Motion Model for Robust Pose Estimation"