Yezhou Yang

New methods simplify personalized image creation from text, enhancing efficiency.

A new approach to reduce bias in AI models and improve predictions.

This study focuses on enhancing spatial accuracy in text-to-image generation.

New method aims to improve the safety of text-to-image generation.

A new framework enhances how models understand spatial relationships in images.

New model improves accuracy and detail in generating 3D shapes from text.

A new dataset aims to improve traffic signal management using real images and diverse scenarios.

Event cameras capture changes in scenes, revolutionizing how we interpret visual data.

Research aims to improve safety at roundabouts by predicting dilemma zones.

This model improves analysis of unevenly spaced time-based data across various fields.

TROPE enhances image captions by focusing on object parts for better detail.

This study highlights the vital role of precise captions in model training.

Explore how RFMs transform image generation in creative fields.