Yanwei Fu

A new framework improves image creation from text and styles.

A new method enhances example selection for visual learning tasks.

Innovative approach to creating clear street views from in-car video footage.

VividPose improves human image animations, making them more realistic and engaging.

This method adjusts object representation slots based on image complexity.

TemporalStory improves image generation for storytelling by enhancing coherence and context.

A new approach enhances the connection between images and text for better interpretation.

Exploring new methods to improve neural surface reconstruction using diverse features.

A new method for seamless 3D editing using multi-view images.

FitDiT improves online shopping with realistic fitting and textures.

MVGenMaster revolutionizes 3D content creation with innovative image generation techniques.

Robots learn to grasp objects quickly with SparseGrasp, using minimal images.