Our research explores the intersection of multimodal perception, embodied intelligence, and interactive systems — aiming to bridge AI with human cognition and action through vision, interaction and co-embodiment.

• Visual & Multimodal Analysis, Enhancement, and Generation

• Egocentric (First-Person) Perception

• Interactive Vision-Language Large Models

• Gaze Estimation & Visual Interaction

• Multimodal Interaction via XR

• Cognitive and Psychological Computing

• The collaborative embodiment of a unified cognitive entity formed by Human Intelligence (HI) and Artificial Intelligence (AI)

• Intelligent Hardware (AI Glasses, AR/VR/MR, Robots)