PNU CVSP Lab develops scalable and reliable spatial multimodal AI for understanding, generating, and interacting with the physical world.
Our research focuses on:
-
3D/4D world modeling and dynamic Gaussian splatting
-
Efficient and reliable vision-language models
-
Embodied spatial intelligence for robots, media, and real-world applications
공경보 pnu-cvsp
정성훈 github
진인환 github
정다운 github
백창우 github
문형주 github
우민수 github
이우진 github
김소현 github
김소형 github
최지훈 github
조범권 github
김우형 github
윤지원 github
윤상연 github