SelfReconCode
SelfReconCode copied to clipboard
Question about avatar generation
Thank you for your great work!!
I have a question about avatar generation in sec 4.4 of your paper. I want to infer with new SMPL pose parameters for avatar generation.
But, I don't know how to do this. I just tried to infer by changing the pose parameter in the existing datasets.py file, but it didn't seem to work this way. Could you please briefly explain how to do it?
I extract the template mesh from the sdf, and bind the mesh with the canonical smpl skeleton and the skinning weights field. Then you can drive the template mesh with new smpl poses.
I extract the template mesh from the sdf, and bind the mesh with the canonical smpl skeleton and the skinning weights field. Then you can drive the template mesh with new smpl poses.
Thank you!!!