Chris Choy

I’m a Senior Research Scientist at NVIDIA working on computer vision, 3D reconstruction, and neural networks. My research focuses on developing algorithms for understanding and reconstructing 3D scenes from various input modalities.

Research Interests

  • 3D Computer Vision: 3D scene understanding, foundation models, 3D reconstruction, geometric deep learning
  • Applications: Robotics, autonomous systems, AR/VR

🌟 Featured Research

Mosaic3D: Foundation dataset and model for open-vocabulary 3D segmentation
MinkowskiNet: Sparse 3D CNNs for efficient processing of large-scale point clouds
FCGF: Fast and accurate 3D feature extraction for scene understanding
Neural Radiance Fields: Advancing view synthesis and 3D reconstruction

Background

I completed my PhD at Stanford University, where I worked on various aspects of computer vision and machine learning. My dissertation, titled “High-dimensional Convolutional Neural Networks for 3D Perception”, focuses on learning-based approaches to tackle challenges in 3D perception across three main categories: reconstruction, representation learning, and registration. The work introduces sparse tensor networks for efficient processing of spatially sparse data and proposes high-dimensional convolutional networks operating in 4D to 32D spaces for geometric pattern recognition. A key contribution is the development of 4D spatio-temporal convolutional neural networks and fully convolutional geometric features that can effectively capture both local and global 3D structures for correspondences and registration tasks. You can find the full thesis at the Stanford Digital Repository here.

Education

  • Ph.D. in EE at Stanford University
  • M.S. in EE at Stanford University
  • B.S. in EE at Korea Advanced Institute of Science and Technology
    • Summa Cum Laude

Honors and Awards

  • 1st place of the 2019 ScanNet Semantic Segmentation Challenge
  • Stanford SystemX FMA Fellowship
  • The Korea Foundation for Advanced Studies, Ph.D. Fellowship
  • The 5th Presidential Science Fellowship for Undergraduate Study
  • Korea Advanced Institute of Science and Technology EECS Merit Scholarship for Best Performance
  • Korea Advanced Institute of Science and Technology Alumni Chairman’s Award
  • Korea Science Olympiad, Junior High, Gold Medal

Patents

  • Universal correspondence network, US Patent App. 10/115,032
  • Systems and Methods for Semantic Segmentation of 3D Point Clouds, US Patent App. 16/155,843
  • Computer-based techniques for learning compositional representations of 3D point clouds, US Patent 11,869,149
  • Action-conditional implicit dynamics of deformable objects, US Patent 12,165,258
  • Using neural networks to perform object detection, instance segmentation, and semantic correspondence from bounding box supervision, US Patent App. 17/177,068
  • Scene reconstruction from monocular video, US Patent App. 18/524,803
  • Sparse voxel transformer for camera-based 3d semantic scene completion, US Patent App. 18/515,016

Contact