Garment Deformation
Garment deformation research focuses on accurately modeling and simulating how clothing moves and deforms in response to human body motion. Current efforts concentrate on developing robust and efficient methods for 3D garment reconstruction and animation, often employing neural networks (including implicit neural representations, graph neural networks, and diffusion models) to capture complex interactions between clothing and the body. This research is crucial for advancing virtual try-on technologies, creating realistic digital avatars for gaming and film, and improving robotic manipulation of clothing. The development of high-quality datasets of real-world clothing and motion is also a significant focus, enabling more accurate and generalizable models.
Papers
DiffMimic: Efficient Motion Mimicking with Differentiable Physics
Jiawei Ren, Cunjun Yu, Siwei Chen, Xiao Ma, Liang Pan, Ziwei Liu
CloSET: Modeling Clothed Humans on Continuous Surface with Explicit Template Decomposition
Hongwen Zhang, Siyou Lin, Ruizhi Shao, Yuxiang Zhang, Zerong Zheng, Han Huang, Yandong Guo, Yebin Liu