CVPR2025
X-Dyna: Expressive Dynamic Human Image Animation
Di Chang, Hongyi Xu, You Xie, Yipeng Gao, Zhengfei Kuang, Shengqu Cai, Chenxu Zhang, Guoxian Song, Chao Wang, Yichun Shi, Zeyuan Chen, Shijie Zhou, Linjie Luo, Gordon Wetzstein, Mohammad Soleymani
Abstract
w w w w w wi i it t th h h h h Dynamic Background D D Dy yn na a am m mi ic c B B B Ba a ac ck k k k kg g gr r ro o ou un n n n nd d d d Static Human S S St t ta a a at t t ti i i ic c c c H H H H H H Hu u u u um m m m m ma a an n n n n with w w w w w w wi i it t t th h h h h Dynamic Background D D D D Dy yna am m m mic c c B B B Ba a ac c ck k k kg g gr ro o ou un n n n nd d d d * Equally contributed as second authors ural scene dynamics from a diverse blend of human and scene videos. Comprehensive qualitative and quantitative evaluations demonstrate that X-Dyna outperforms state-ofthe-art methods, creating highly lifelike and expressive animations.