3 papers
cs.CV2025
MoSA: Motion-Coherent Human Video Generation via Structure-Appearance Decoupling
Haoyu Wang, Hao Tang, Donglin Di +5
Existing video generation models predominantly emphasize appearance fidelity while exhibiting limited ability to synthesize complex human motions, such as whole-body movements, lon…
cs.CV2024
MV-VTON: Multi-View Virtual Try-On with Diffusion Models
Haoyu Wang, Zhilu Zhang, Donglin Di +2
The goal of image-based virtual try-on is to generate an image of the target person naturally wearing the given clothing. However, existing methods solely focus on the frontal try-…
cs.CV2023
Self-Supervised High Dynamic Range Imaging with Multi-Exposure Images in Dynamic Scenes
Zhilu Zhang, Haoyu Wang, Shuai Liu +3
Merging multi-exposure images is a common approach for obtaining high dynamic range (HDR) images, with the primary challenge being the avoidance of ghosting artifacts in dynamic sc…