Qixing Huang

A look into improving human motion generation from text descriptions.

VideoMV transforms 2D images and text into fast, consistent 3D visuals.

Introducing a method for improved image matching across diverse visual data.

A new system enables 3D model creation using single real-world images.

A fresh approach to 3D shape manipulation without overlaps.

Explore new methods in generating realistic human movements from text descriptions.

PPLNs enhance event camera data processing for improved machine vision capabilities.

Researchers enhance 3D reconstructions using synthetic data for better results.