search

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".
137 stars 19 forks source link

5.1.2 Model-based Approaches #18

Closed lihuiGZHU closed 7 months ago

lihuiGZHU commented 1 year ago

I have a few questions about model-based methods which use 3D skeletons as the input. 1.Does your code reflect the use of posegait to train 3Dskeleton data？ 2.The 3D skeleton data in Gait3D is under which folder in your data？

JinkaiZheng commented 1 year ago

Hi~ Thank you for your question.

  1. We reproduce PoseGait based on the description of the PoseGait paper, combined with the partial code published by the author (https://github.com/RijunLiao/PoseGait)
  2. The 3D skeleton data in Gait3D is under 3D SMPLs NPZ. Specifically, the 3D skeleton used in our paper is j3d_smpl24 in 3D SMPL NPZ.
lihuiGZHU commented 1 year ago

Thank you for your reply！

lihuiGZHU commented 1 year ago

I also want to ask you a question, do you have a backup of the code for reproducing posegait using Gait3D's 3Dskeleton data as input, and if so, can you send me a copy, because it will greatly reduce my workload. This is my email lihui0627@e.gzhu.edu.cn, thank you very much! ! !

JinkaiZheng commented 1 year ago

I'm sure you'll be able to do it yourself as described in the PoseGait paper, and it won't take you too much time, so good luck.