Zongxin Yang

TransHuman offers a new method for realistic human video rendering.

A new approach improves object segmentation in video using audio-visual integration techniques.

GD-NeRF tackles image blurriness in novel view synthesis.

SIAF improves video segmentation with user-friendly multi-frame interactions.

A new method improves the creation of emotionally expressive talking-head videos.

New methods enhance the creation of multiple objects in images with improved accuracy.

Exploring the risks and implications of replication in visual diffusion model technology.

A new framework enhances the alignment of sounds and visuals in videos.