Shu Kong, Ph.D.

[GitHub] [Google Scholar] [ORCID] [LinkedIn] [X]
Home Publication Datasets Group Contact

I lead the Visual Intelligence Lab. My research vision is to establish the foundations of Lifelong Visual Intelligence, enabling machines to continously learn from the open world, domain experts, and individual users through visual perception and interaction. To realize this vision, I drive three strategic thrusts:

  • Open-World Vision aims to build robust, generalizable, fair, and trustworthy Visual Intelligence systems for real-world open environments. My ICCV'21 paper on this research thrust was recognized for Best Paper / Marr Prize. See my CVPR'24 paper to learn how the Open-World philosophy demystifies inherent biases present in modern AI models.
  • AutoExpert operationalizes domain expertise to make specialized, expertise-intensive applications more accessible. I build intelligent systems for palynology, paleoecology, and evolutionary biology. This includes my PNAS paper featured by the U.S. NSF, which noted that the research "opens a new era of fossil pollen research". While this research thrust can be termed as AI for Science by today's standards, my recent NeurIPS'26 paper formulates AutoExpert for autonomous driving from an industrial perspective.
  • Personalized Visual Intelligence enables systems to learn from individual users and deliver tailored services to everyone. I build systems (e.g., AR/VR wearables and assistive robots) that assist individuals with personal item retrieval [ref1, ref2] and enhance their daily interactions in home environments [ref1, ref2].

I am an Assistant Professor in the Department of Artificial Intelligence, affiliated with the Department of Computer Science and the Institute of Artificial Intelligence and Brain Sciences at the University of Macau. I am also a research affiliate with the Okinawa Institute of Science and Technology. I previously held research and faculty positions at Carnegie Mellon University and Texas A&M University. I completed my postdoc training at the Robotics Institute of Carnegie Mellon University, supervised by Deva Ramanan. I received my Ph.D. in Computer Science from the University of California, Irvine, advised by Charless Fowlkes.

For students who are interested in joining our lab, please read these tips. PhD students will be fully funded with fellowships.

Email contact

  • related to UM: skong [at] um [dot] edu [dot] mo
  • related to OIST: shu.kong [at] oist [dot] jp
  • others: aimerykong [at] gmail [dot] com
  • related to TAMU: shu [at] tamu [dot] edu

Teaching

Workshops

Professional services

Updates