Bohan Zhuang

ModaVerse simplifies how we transform and interpret various types of data.

T-Stitch combines models for faster, high-quality image generation.

Introducing Motion Mamba, a fast method for realistic human motion generation.

MVSplat revolutionizes 3D reconstruction using sparse images with efficient techniques.

A new method to improve memory usage in language models.

A new method to create long, smooth human movements from text prompts.

Introducing McCaD: A new approach to high-quality MRI image synthesis.

Create stunning 3D images from just a few photos effortlessly.

AbilityLens standardizes evaluation for multimodal large language models.

VisionFuse improves AI's understanding of images through collaboration of models.