Chris Choy

Papers

Total Citations

H-Index

About

Chris Choy is a leading researcher in 3D computer vision and deep learning, best known for pioneering work in efficient scene understanding and neural field representations. His research focuses on monocular scene reconstruction, 3D point cloud processing, and sparse convolutional networks, with major contributions to making 3D perception both accurate and computationally practical. His highly influential work on "MinkowskiEngine" introduced sparse tensor networks that revolutionized 3D deep learning, enabling real-time processing of large-scale point clouds—a breakthrough that has garnered thousands of citations and become a standard tool in the field. Choy's recent paper on "Fast Monocular Scene Reconstruction with Global-Sparse Local-Dense Grids" (2023) addresses a long-standing challenge in augmented reality and robotics: reconstructing indoor scenes from single images. By combining global sparsity with local density, his method achieves state-of-the-art surface reconstruction quality while maintaining real-time performance. His work has been recognized with multiple best paper awards and is widely adopted in autonomous driving, robotics, and AR applications. Choy's research continues to push the boundaries of what's possible in 3D scene understanding, making him a pivotal figure in modern computer vision.