alvinliu0 / HA2G

[CVPR 2022] Code for "Learning Hierarchical Cross-Modal Association for Co-Speech Gesture Generation"
https://alvinliu0.github.io/projects/HA2G
GNU General Public License v3.0
127 stars 9 forks source link

About reviewing proj_joints on original images #14

Open robingg1 opened 1 year ago

robingg1 commented 1 year ago

Hi, thanks for your outstanding work ! However I have some problems about the visualization part for the proj_joints extracted from expose.

mm = clip['3d'][i]['focal_length_in_mm']
px = clip['3d'][i]['focal_length_in_px']
proj_joints = proj_joints * px * 10 / mm
proj_joints += center

Here is your code for visualize proj_joints into the original images, I am confused about this code, why it needs to multiple 10 for proj_joints*px. Are there any actual means of this 10 or can you offer some reference for this part ? Thanks !