Thanh-Tung (Tony) Le

UCI - Computer Science
6210 Donald Bren Hall
Irvine, CA 92697-3425
I am on job market and am actively seeking Research Scientist/Research Internship positions. If you are interested in my profile, please feel free to contact me via email at thanhtul[at]uci[dot]edu or my LinkedIn page.
I am a fourth-year PhD candidate in Computer Science at the University of California, Irvine, advised by Professor Xiaohui Xie. My research focuses on Generative Models and 3D Vision, with a broader interest in developing robust Multimodal AI systems. I aim to create general-purpose World Models that can learn from rich visual data, enabling interactive and dynamic environments.
Research interests:
- Generative models
- Diffusion Models, Flow Matching, Autoregressive, Masked Autoregressive Models, Hybrid Generative Approaches.
- Video Generation, Diffusion Distillation, Generative Novel Views Synthesis, Diffusion-based Dense Estimation.
- 3D vision
- Novel View Synthesis, 3D Shape Analysis and Representations, Neural Rendering (NeRF and 3DGS-based).
- 2.5D Scene Representation: Depth/Normals Estimation.
- Efficient Optimal Transport for 3D Shape Learning.
news
Aug 18, 2025 | One paper is accepted by NeurIPS 2025. |
---|---|
Aug 03, 2025 | I advance to PhD candidate and receive En route Master’s in Computer Science. |
Aug 01, 2025 | One paper is accepted by ICCV Findings 2025. |
Jun 16, 2025 | I am honor to receive research internship offer from NVIDIA and Google for Summer 2025. I join Google as a Student Researcher. |
Sep 01, 2024 | One paper is accepted by BMVC 2024 (Oral). |
Jun 23, 2024 | I join Meta Reality Labs as a Research Scientist Intern. |
Jan 21, 2024 | One paper is accepted by CVPR 2024. |
Jan 04, 2024 | One paper is accepted by ICLR 2024. |
Oct 19, 2023 | Two papers are accepted by WACV 2024. |
Sep 11, 2022 | I am honored to received Dean’s Award from Donald Bren School of Information and Computer Sciences, UC Irvine. |
Sep 02, 2022 | I start my Ph.D. journey at UC Irvine! |
selected publications
- NeurIPSUMAMI: Unifying Masked Autoregressive Models and Deterministic Rendering for View SynthesisIn The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025
- CVPRIntegrating efficient optimal transport and functional maps for unsupervised shape correspondence learningIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024
- ICLRDiffeomorphic mesh deformation via efficient optimal transport for cortical surface reconstructionThe Twelfth International Conference on Learning Representations, 2024
- BMVCHybrid-csr: Coupling explicit and implicit shape representation for cortical surface reconstructionThe 35th British Machine Vision Conference (BMVC 2024), 2024
- ICCV-FindingsGeoDiff: Geometry-Guided Diffusion for Metric Depth EstimationIn First Workshop on the Findings of the International Conference on Computer Vision 2025, 2025