Ying Shan

ReVideo enables precise video editing by controlling both content and motion.

Learn how MOFA-Video transforms still images into engaging animations.

CV-VAE improves video generation efficiency and quality in existing models.

A novel dataset and method enhance video grounding for complex narratives.

A new method for accurate depth estimation from video sequences.

A project focused on enhancing image generation through advanced techniques and models.

A new benchmark pushes boundaries in video understanding by focusing on event-level tasks.

Discover how RF technology transforms image and video creation.

NovelGS uses sparse images to create impressive 3D models efficiently.

DOGE simplifies document interaction, making information accessible and understandable.

Moto uses video analysis to teach robots complex movements efficiently.

Discover how Divot transforms video comprehension and generation.

FreeSplatter creates detailed 3D models from random images without camera data.

A new model blends music and AI, creating innovative tunes.

ColorFlow breathes life into black and white images, ensuring vibrant consistency.

A new tool simplifies creating 3D models, boosting efficiency for artists and designers.