4DHands: Reconstructing Interactive Hands in 4D with Transformers
Dixuan Lin1,2, Yuxiang Zhang2, Mengcheng Li2, Yebin Liu2,
Wei Jing3, Qi Yan3, Qianying Wang3*, Hongwen Zhang1*
1Beijing Normal University, 2Tsinghua University, 3 Lenovo
Code
4DHands can robustly recover interactive hand meshes and their relative movement from monocular inputs.
Results - Monocular Video Input
4DHands outperforms other interactive hand reconstruction methods in real-world videos.
Results - Single Image Input
4DHands has robust performance in complex cases.
Results - Complex Interactions
4DHands has robust performance in complex cases.