Introduction
The Computer Vision & Signal Processing (CVSP) Lab focuses on developing intelligent systems that understand and interact with the real world. Our research goes beyond conventional perception, aiming to build multimodal world models that integrate vision, language, and spatial understanding. We study how AI systems can reason about complex environments and enable robust decision-making in real-world scenarios.
Our main research directions include:
1.
Multimodal foundation models (vision, language, and audio) for grounded real-world understanding,
2.
3D/4D scene representation and dynamic world modeling for spatio-temporal reasoning,
3.
Embodied intelligence and action-aware perception for bridging perception and decision-making.
We aim to develop AI systems that move beyond passive perception, enabling machines not only to see, but to understand, reason, and act.
.png&blockId=2f2abcc9-0e34-8089-ae0a-ccc7a9740049)
.png&blockId=2f2abcc9-0e34-8066-91bc-e83299b2685c)
