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".
133 stars 19 forks source link

ROMP SMPL transfer #8

Closed ThomasNing closed 2 years ago

ThomasNing commented 2 years ago

When I try to use the ROMP to generate out the 3D mesh, I detect there is a version conflict with the ROMP used by SMPLGait. Could I know which version of the ROMP the SMPLGait used? In this way I could use the SMPLGait to run on other ReID dataset.

Gait3D commented 2 years ago

Dear Thomas, The version of ROMP we used is V1.0, but the latest version should be compatible.

Best regards, The group of Gait3D

Thomas Ning @.***> 于2022年8月2日周二 13:40写道：

When I try to use the ROMP to generate out the 3D mesh, I detect there is a version conflict with the ROMP used by SMPLGait. Could I know which version of the ROMP the SMPLGait used? In this way I could use the SMPLGait to run on other ReID dataset.

— Reply to this email directly, view it on GitHub https://github.com/Gait3D/Gait3D-Benchmark/issues/8, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYEL7JR2BPLDFRTJ73KRVE3VXCYE7ANCNFSM55JZV5ZA . You are receiving this because you are subscribed to this thread.Message ID: @.***>

ThomasNing commented 2 years ago

I use the latest version of the ROMP. Unfortunately, the ROMP team changes their .npz data output format, with different names of the result and also the matrix shape size of the poses. So the preprocess_smpl.py document failed to preprocess the latest romp output. What I can do is to modify the preprocess code to align the new format of the romp output.

JinkaiZheng commented 2 years ago

Okay, good luck!

ZziTaiLeo commented 1 year ago

I use the latest version of the ROMP. Unfortunately, the ROMP team changes their .npz data output format, with different names of the result and also the matrix shape size of the poses. So the preprocess_smpl.py document failed to preprocess the latest romp output. What I can do is to modify the preprocess code to align the new format of the romp output.

Have you solved the problem? Can you share the code of data preprocessing after ROMP.