Scholar
Ang Cao
Google Scholar ID: HtD-aVUAAAAJ
University of Michigan
Citations & Impact
Citations
1,166
H-index
7
i10-index
7
Publications
13
Co-authors
7
list available
Publications
Resume (English only)
Academic Achievements
- From Thousands to Billions: 3D Visual Language Grounding via Render-Supervised Distillation from 2D VLMs, ICML 2025
- Probing Visual Language Priors in VLMs, ICML 2025 (equal contribution)
- Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass, CVPR 2025
- Meta 3D Gen, Tech report, 2024
- Lightplane: Highly-Scalable Components for Neural 3D Fields, 3DV 2025
- DreamGaussian4D: Generative 4D Gaussian Splatting, ArXiv 2023
- Text2Room: Extracting Textured 3D Meshes from 2D Text-to-Image Models, ICCV 2023 (Oral, equal contribution)
- HexPlane: A Fast Representation for Dynamic Scenes, CVPR 2023
- FWD: Real-time Novel View Synthesis with Forward Warping and Depth, CVPR 2022
- Inverting and Understanding Object Detectors
Background
- Final-year Ph.D. student (2020–present) in Computer Science and Engineering (CSE) at the University of Michigan, Ann Arbor
- Research focuses on generative models and 3D vision, especially world modeling and 3D world creation
- Key research includes: 3D/4D reconstruction and generation from 2D data (e.g., HexPlane, FWD, Lightplane)
- Injecting world knowledge into 3D/4D via generative/foundation models such as VLMs and diffusion models (e.g., Text2Room, LiftGS)
- Also explores novel ideas in generative models (e.g., ViLP)
