mks0601 / I2L-MeshNet_RELEASE

Official PyTorch implementation of "I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Image", ECCV 2020

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

H36m preprocessing file Require

MooreManor opened this issue · comments

Hello. Thanks for your great work!
It seemed that I missed the preprocessing file of changing videos in h36m to images whose format is like s_01_act_02_subact_01_ca_01*.jpg and generating corresponding JSON files like Human36M_subject*_data.json.
I have the video of those images but I don’t have the preprocessing file to change the video to images. Can you offer the preprocessing files? Thanks!

You can download pre-processed files in here

@mks0601 Would you be pleased to offer the parsing code of that? Because I have the source videos and annotations of that. Downloading all files from Google Drive is slow and may have download failure for a huge file. Thanks!

Maybe you sent the mail? I have replied. Thanks!

Okay! That's fine. I got it. Thanks a lot!

@mks0601 Is this the parsing code to generate pics? H36M-Toolbox