1 paper
Jisu Shin, Junmyeong Lee, Seongmin Lee +4
We present a novel framework for reconstructing animatable human avatars from multiple images, termed CanonicalFusion. Our central concept involves integrating individual reconstru…