3D hand reconstruction with both shape and appearance from an RGB image

Xiaoyun Chang, Wentao Yi, Xiangbo Lin, Yi Sun

Abstract: Highlights•We present a method to estimate a 3D realistic hand with both shape and appearance.•We propose a structural implicit 3D hand representation with part semantics.•We present a self-supervised hand appearance synthesis network.•The superiority of our method is demonstrated in extensive experiments.