MHFormer issues

Missing reference paper that has better results on `Human3.6M` than `MHFormer`?

1

The paper "[Occlusion-Aware Networks for 3D Human Pose Estimation in Video](https://openaccess.thecvf.com/content_ICCV_2019/papers/Cheng_Occlusion-Aware_Networks_for_3D_Human_Pose_Estimation_in_Video_ICCV_2019_paper.pdf)"[1] published in ICCV 2019 achieves 42.9 MPJPE(protocol 1) and 32.8 P-MPJPE(protocol 2) while your paper[2] achieves 43.0 MPJPE and...

sieu-n

IndexError: list index out of range?

When i tried to train a new model, this error occurred: File "main.py", line 117, in model_path = sorted(glob.glob(os.path.join(opt.previous_dir, '*.pth')))[1] IndexError: list index out of range I wonder why and...

TomCatOnline

Can you share the training and testing codes of the 3dhp dataset?

2

I used the script provided by the P-STMO model(https://github.com/paTRICK-swk/P-STMO) to train and test the MHFormer model on the 3dhp dataset. The average value of PCK and AUC indexes obtained was...

ClimberY

How to get the transition of root?

4

It looks like there is no output of transition of root joint?

lucasjinreal

Some weird codes in common/generator.py

Line 18: "assert poses_3d is None or poses_3d[key].shape[0] == poses_3d[key].shape[0]" What's the meaning of "poses_3d[key].shape[0] == poses_3d[key].shape[0]" here? Line 22: "augment_vector = np.full(len(bounds - 1), False, dtype=bool)" What's the difference...

iskaiwen

Is it possible to use YOLOv7 ?

2

非常棒的工作！我正在拼命学习你们的研究并且尝试将其应用于实际领域。我想问的是， 1. output文件夹下的'input_2D'文件夹内是否已经是3D 人体姿势估计的结果（也就是你们论文中所描述的主要工作生成的结果）？因为我想将其用在优化虚拟人物动作中。 2. 目前使用的是略显“古老”的YOLOv3和体积很大的HRNet来工作，是否有可能将其换成新的YOLOv7和Lite-HRNet，以加快识别姿态的速度，并最终实现实时识别和多目标同时识别？如果可行的话，能否给我讲一下修改方法？例如如何生成“ YOLOv7.weights”文件，以及能否直接在github仓库中下载Lite-HRNet来替换使用？请原谅我作为一个cv领域的初学者可能提问过于小白，感谢您的耐心 Great Great Job! I'm trying my best to study your research and try to apply it in real world....

Arkitect-z

Add Docker environment and web demo

1

Hey [@Vegetebird ](https://github.com/Vegetebird)! 👋 This pull request makes it possible to run your model inside a Docker environment, which makes it easier for other people to run it. We're using...

vivcheng01