PHAI

Faculty

Dr. Zhiwen Fan is currently an Assistant Professor in the Department of Electrical and Computer Engineering (ECE), Texas A&M University, College Station, TX, USA.

He leads research on computer vision, generative models, and robotic perception at Texas A&M.

Email Homepage Google Scholar LinkedIn Twitter

Research Interest

If you are here to seek "TL;DR"...

Question: what should I read, if I just want to have a very quick glance at what Zhiwen Fan's group is currently focusing on?
Answer: Prof. Fan's research goal is to build spatial foundations for embodied and XR intelligence by developing world modeling systems from multimodal physical data. His research spans real-time spatial reasoning and world modeling with vision-language systems, 3D/4D reconstruction and generation, digital humans, and AI for scientific and biomedical imaging. Below are six recent papers that reflect these directions. This list will be updated over time:

PhD Positions Available for Spring 2026, and Fall 2026; Welcome Research Interns for Spring/Summer 2026.

Academic Activities and Services

Dr. Zhiwen Fan received his Ph.D. degree in Electrical and Computer Engineering from The University of Texas at Austin, Austin, TX, USA, advised by Prof. Zhangyang "Atlas" Wang. Dr. Fan's research focuses on computer vision, generative models, and robotic perception. He has co-authored 50+ papers with 5,000+ citations, including one paper with 1,000+ citations, with publications in venues such as NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, and IEEE TPAMI, as well as related publications in IPMI, IROS, and ICCAD.

He has received the Best Paper Award at ACM MM 2025's Multimodal Foundation Models for Spatial Intelligence Workshop, the Best Paper Award at CVPR 2025's AI for Content Creation Workshop, the Qualcomm Innovation Fellowship (North America) 2022, 3rd Place in the University Demo Best Demonstration at DAC 2022, and 3rd Place in the ICCV 2025 COGS Challenge for Compact 3D Representation. He has served as an Area Chair for ICLR 2025, NeurIPS 2025, and ICASSP 2026, an Associate Editor for IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), and an organizer of workshops on 3D learning, embodied AI, agentic AI, and scientific imaging at multiple top-tier AI conferences.

News

[Feb 2026]

3 CVPR main track papers (VLM-3R, SpatialStack, HumanNOVA) and 3 CVPR findings papers accepted.

[Jan 2026]

1 paper (Robot Planning with Formal Methods Feedback) was accepted by ICRA 2026.
Gift from Meta, thanks!

[Dec 2025]

Four workshops will be held at CVPR 2026 in Denver, covering these topics: End-to-End 3D Learning; Foundation Models Meet Embodied Agents; Multi-Agent Embodied Intelligent Systems in the Agentic AI Era: Opportunities, Challenges, and Future Directions; and 3D Geometry Generation for Scientific Computing.

[Nov 2025]

Named as Top Area Chair for NeurIPS 2025.

[Oct 2025]

Our paper VLM-3R received the Best Paper Award at ACM MM 2025 Multimodal Foundation Models for Spatial Intelligence Workshop.
At End-to-End 3D Learning (E2E3D) workshop, we awarded one Best Paper Award (CSG-Fusion) and one Best Demo Award (PointSeg). See details here.
We won 3rd place in the ICCV 2025 COGS Challenge for Compact 3D Representation.

[Sep 2025]

I will serve as the Area Chair for ICLR 2026.
2 papers (DynamicVerse, Martian World Model) were accepted by NeurIPS 2025 and NeurIPS 2025 D&B track.

[Jun 2025]

Our paper VideoLifter received the Best Paper Award at CVPR 2025’s AI for Content Creation Workshop.
2 papers (CryoFastAR, X2-Gaussian) were accepted by ICCV 2025.
We are organizing End-to-End 3D Learning workshop at ICCV 2025.

[May 2025]

I will serve as the Area Chair for NeurIPS 2025.