facebookresearch / goliath

Goliath Dataset and Official PyTorch Implementation of RelightableHands, Relightable Gaussian Codec Avatars, and Driving-Signal Aware Full-Body Avatars.
Other
174 stars 14 forks source link

Clarification on URHand personalization #21

Open zc-alexfan opened 6 days ago

zc-alexfan commented 6 days ago

Hi,

Thanks for sharing the amazing work. We are trying to compare with URHand for personalization.

However, we want to see if this codebase is sufficient for the comparison in our setting. In particular, given one or more RGB images of a right hand, we would like to use URHand to learn a personalized hand avatar. After this avatar is learnt, given an environment map in HDRI (e.g., in exr) format, and a set of MANO poses, can this codebase yield the relit hand images?

Thanks!

una-dinosauria commented 2 days ago

The codebase does not support MANO currently; we provide our own registration hand topology instead, so you'd have to convert between the two.