Anh N. Thai (Ant)

I am a Senior AI Multimodal Researcher at Dolby Laboratories. I was a PhD student in the School of Interactive Computing (IC) at Georgia Institute of Technology, advised by Professor James M. Rehg and Professor Judy Hoffman. My research centers around understanding 3D scenes and objects.

Prior to PhD, I graduated with a B.S. in Computer Science from Georgia Tech.

I’m interested in analog photography and playing musical instruments, especially the guitar. I love water activities and I draw sometimes.

Contacts

Publications

SplatTalk: 3D VQA with Gaussian Splatting

Anh Thai, Songyou Peng, Kyle Genova, Leonidas Guibas, Thomas Funkhouser. ICCV 2025 (Poster)

Symmetry Strikes Back: From Single-Image Symmetry Detection to 3D Generation

Xiang Li, Zixuan Huang, Anh Thai, James M. Rehg. CVPR 2025 (Highlight)

3 × 2: 3D Object Part Segmentation by 2D Semantic Correspondences

Anh Thai, Weiyao Wang, Hao Tang, Stefan Stojanov, James M. Rehg, Matt Feiszli. ECCV 2024 (Poster)

Leveraging Object Priors for Point Tracking

Bikram Boote, Anh Thai, Wenqi Jia, Ozgur Kara, Stefan Stojanov, James M. Rehg, Sangmin Lee. ECCVW (Oral)

ZeroShape: Regression-based Zero-shot Shape Reconstruction

Zixuan Huang*, Stefan Stojanov*, Anh Thai, Varun Jampani, James M. Rehg. CVPR 2024 (Poster)

Low-shot Object Learning with Mutual Exclusivity Bias

Anh Thai, Ahmad Humayun, Stefan Stojanov, Zixuan Huang, Bikram Boote, James M. Rehg. NeurIPS 2023 (Poster)

ShapeClipper: Scalable 3D Shape Learning via Geometric and CLIP-based Consistency

Zixuan Huang, Varun Jampani, Anh Thai, Yuanzhen Li, Stefan Stojanov, James M. Rehg. CVPR 2023 (Poster)

Learning Dense Object Descriptors from Multiple Views for Low-shot Category Generalization

Stefan Stojanov, Anh Thai, Zixuan Huang, James M. Rehg. NeurIPS 2022 (Poster)

The Surprising Positive Knowledge Transfer in Continual 3D Object Shape Reconstruction

Anh Thai, Stefan Stojanov, Zixuan Huang, Isaac Rehg, James M. Rehg. 3DV 2022 (Poster)

Planes vs. Chairs: Category-guided 3D shape learning without any 3D cues

Zixuan Huang, Stefan Stojanov, Anh Thai, Varun Jampani, James M. Rehg. ECCV 2022 (Poster)

Using Shape to Categorize: Low-Shot Learning with an Explicit Shape Bias

Stefan Stojanov, Anh Thai, James M. Rehg. CVPR 2021 (Poster)

3D Reconstruction of Novel Object Shapes from Single Images

Anh Thai*, Stefan Stojanov*, Vijay Upadhya, James M. Rehg. 3DV 2021 (Oral)

Incremental Object Learning from Contiguous Views

Stefan Stojanov, Samarth Mishra*, Anh Thai*, Nikhil Dhanda, Ahmad Humayun, Chen Yu, Linda B. Smith, James M. Rehg. CVPR 2019 (Oral - Best paper finalist)

(*Equal contribution)

News

• June 2025 Our work SplatTalk: 3D VQA with Gaussian Splatting was accepted for poster publication at ICCV 2025
• March 2025 I defended my PhD dissertation
• February 2025 I started my full-time job at Dolby Laboratories as Senior AI Multimodal Researcher
• February 2025 Our work Symmetry Strikes Back: From Single-Image Symmetry Detection to 3D Generation was accepted for highlight publication at CVPR 2025
• July 2024 Our work 3×2: 3D Object Part Segmentation by 2D Semantic Correspondences was accepted for publication at ECCV 2024
• February 2024 I accepted Summer 2024 student researcher offer at Google Deepmind
• February 2024 Our work ZeroShape: Regression-based Zero-shot Shape Reconstruction was accepted for poster publication at CVPR 2024
• September 2023 Our work Low-shot Object Learning with Mutual Exclusivity Bias was accepted for publication at NeurIPS 2023
• February 2023 I accepted Summer 2023 research internship offer at Meta Research (FAIR)
• February 2023 Our work ShapeClipper: Scalable Learning 3D Shape Learning via Geometric and CLIP-based Consistency was accepted for publication at CVPR 2023
• September 2022 Our work Learning Dense Object Descriptors from Multiple Views for Low-shot Category Generalization was accepted for poster publication at NeurIPS 2022
• August 2022 Our work The Surprising Positive Knowledge Transfer in Continual 3D Object Shape Reconstruction was accepted for poster publication at 3DV 2022
• July 2022 Our work Planes vs. Chairs: Category-guided 3D shape learning without any 3D cues was accepted for poster publication at ECCV 2022
• April 2022 I was selected to participate in ICVSS 2022 in Sicily, Italy
• October 2021 Our work 3D Reconstruction of Novel Object Shapes from Single Images was accepted for oral publication at 3DV 2021
• February 2021 Our work Using Shape to Categorize: Low-Shot Learning with an Explicit Shape Bias was accepted for poster publication at CVPR 2021
• January 2021 I accepted Summer 2021 research internship offer at Facebook Reality Labs Research
• August 2019 I am officially a first-year PhD student at Georgia Tech
• April 2019 I will be attending Georgia Tech starting from Fall 2019 to pursue a PhD in Computer Science
• March 2019 Our work Incremental Object Learning from Contiguous Views was accepted for oral presentation at CVPR 2019