Hand4Whole_RELEASE
Hand4Whole_RELEASE copied to clipboard
Official PyTorch implementation of "Accurate 3D Hand Pose Estimation for Whole-Body 3D Human Mesh Estimation", CVPRW 2022 (Oral.)
I want to enter an image of (512, 512) or (256,256). Because I am not very clear about the role and relationship of these sizes in the reconstruction process, I...
Hi, when I run agora2coco.py to make a comment file for AGORA, I get the following error. I don't know what the problem is, can you help me? Thanks a...
Hello Gyeongsik, I have encountered issues while downloading the MPI_INF_3DHP dataset from the google drive link you have provided. The webpage says the daily download limit has exceeded, and I...
Hello, when I use your full-body model for rendering, I found that when my fingers touch, the effect is a certain gap with the imagination, and it will only be...
Hello, can you help me to see what might be wrong? I only used MPII data for training, and then generated the model after the second stage for running the...
I only used MSCOCO for train. Trying the 1st: 1st: pre-train Hand4Whole and run `python train.py --gpu 0-3 --lr 1e-4 --continue` Getting the following error `Traceback (most recent call last):...
Hello author, in the figure 1 shown in the article, '3D wrist rotations with MCP joints', only two vectors of '2x45' are seen in the code regarding the 'pose' parameter...
vis_3d_skeleton(kpt_3d, kpt_3d_vis, kps_lines, filename=None) 1. Is' kpt_3d 'passed in' joint_img '? 2. What is passed in 'kpt_3d_vis',' kps_lines', and where to get it?
Hi, thanks for the great work! I would like to ask if I want to use 3D pose as the input, how should I re-train the model? Basically, I want...
Thanks for your great work! I have a question about face key points names.  In human_models.py, it's mentioned that face key points names can be found in FLAME class....