Abstract: Modeling a human avatar that can plausibly deform to articulations is an active area of research. We present PocoLoco - the first template-free, point-based, pose-conditioned generative ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback