Gait3D / Gait3D-Benchmark

This is the code for the paper "Gait Recognition in the Wild with Dense 3D Representations and A Benchmark. (CVPR 2022)", "Gait Recognition in the Wild with Multi-hop Temporal Switch", and "Parsing is All You Need for Accurate Gait Recognition in the Wild".
133 stars 19 forks source link

Question about 3D SMPL skeleton topology diagram #11

Closed HL-HYX closed 1 year ago

HL-HYX commented 2 years ago

Your work promotes the application of gait recognition in real scenes, can you provide the topology diagram of the SMPL 3D skeleton in Gait3D? Because the specific meaning of the 24 joint points is not stated in your data description document.

JinkaiZheng commented 2 years ago

Hi~ I'm so sorry. Our poor consideration indeed causes this. The following image is what you want.

dc4f04733e45940e4b6aa86b0ebd6634

HL-HYX commented 2 years ago

Thank you very much for your reply!

HL-HYX commented 2 years ago

Hello, sorry to bother you again! Just I have another question: In your ReadMe.txt, there are two types of data framed by the red box in the figure below. From a visualization of your data, I found that j3d_smpl24(24, 3) describes the 3D coordinates of the nodes. So what does pose(72, ) mean? image

JinkaiZheng commented 2 years ago

Yes, _j3dsmpl24 (24, 3) means the 3D keypoints coordinate and pose (72, ) equals 24*3, specifically, is the rotation (axial angle) of each joint around the X/Y/Z axis in the above figure.