Diffusion Based VTON
Diffusion-based virtual try-on (VTON) aims to realistically render clothing onto images of people, enhancing online shopping experiences. Recent research focuses on improving the accuracy and realism of try-on results, particularly for challenging poses and clothing types, by employing advanced diffusion models and addressing issues like semantic inconsistencies and the need for precise masks. Key advancements include incorporating multi-view information, disentangling semantic features for improved garment warping, and developing novel loss functions to enhance detail and accuracy. This field significantly impacts e-commerce by providing more realistic and engaging virtual shopping experiences.
Papers
MC-VTON: Minimal Control Virtual Try-On Diffusion Transformer
Junsheng Luan, Guangyuan Li, Lei Zhao, Wei Xing
VTAO-BiManip: Masked Visual-Tactile-Action Pre-training with Object Understanding for Bimanual Dexterous Manipulation
Zhengnan Sun, Zhaotai Shi, Jiayin Chen, Qingtao Liu, Yu Cui, Qi Ye, Jiming Chen