Jinwei Gu

Senior Research Scientist (NVIDIA)

Learning and Perception Research
NVIDIA Research
2788 San Tomas Expy, Santa Clara, CA 95050
Office: Building Endeavor
Email: FirstnameG AT
NVIDIA Webportal
CV   [My Google Scholar Citation Page]

Recent Research


MapNet: Geometry-Aware Learning of Maps for Camera Localization[PDF][Supplementary Video][Project (coming soon)]

with Samarth, Kihwan, James, and Jan. CVPR 2018.


Reblur2Deblur: Deblurring Videos via Self-Supervised Learning[PDF][Supplementary Video][Project (coming soon)]

with Huaijin, Orazio, Ming-Yu, Ashok, and Jan. ICCP 2018.


Separating Reflection and Transmission Images in the Wild [PDF][Supplementary Video][Project (coming soon)]

with Patrick, Orazio, and Jan. Arxiv 2017.


Learning Affinity via Spatial Propagation Networks [PDF][Supplementary Video][Project (coming soon)]

with Sifei, Shalini, Guangyu, Ming-Hsuan,and Jan. NIPS 2017 (Accepted).


A Lightweight Approach for On-the-Fly Reflectance Estimation[PDF][Supplementary Video][Project (coming soon)]

with Kihwan, Stephen, Pavlo, Matthias, and Jan. ICCV 2017 Oral. arXiv:1705.07162


Dynamic Facial Analysis: From Bayesian Filtering to Recurrent Neural Networks [Project][PDF][Supplementary Video]

with Xiaodong, Shalini and Jan. CVPR 2017


Polarimetric Multi-view Stereo [PDF][Supplementary Document][Video]

with Zhaopeng, Boxin, Ping and Jan. CVPR 2017


Reconstructing Intensity Images from BInary Spatial Gradient Cameras [PDF][Supplementary Document][Video]

with Suren, Orazio, Timo and Jan. CVPR EVW 2017 (Oral)


Deep Learning with Energy-efficient Binary Gradient Cameras [PDF]

with Suren and Orazio. arXiv:1612.00986, 2016


Video Stitching with Spatial-Temporal Content-Preserving Warping[PDF][Video Results]

with Wei. CVPR 2015 Workshop on Computational Cameras and Displays


Previous Research


Exemplar-Based Method for Automatic Visual Editing and Retouching of Fine Art Reproduction [project]

with Jun. CIC 2013


Learning Discriminative Illumination and Filters for BTF Classification[project] [database](with code)

with Chao and Gefei. CVPR 2013


What is the Space of Camera Spectral Sensitivities?[project] [database](with code)

with Jun, Dengyu, and Sabine. IEEE WACV 2013


Classifying Raw Materials with Discriminative Illumination [project] [database]

with Chao. CVPR 2012 (Oral)


Spectral Imaging Under Commonly Available Lighting [project]

with Jun. CVPR 2012 Workshop on Computational Camera and Displays


Multiplexed Illumination for Scene Recovery in the Presence of Global Illumination[project]

with Toshi, Mohit, and Shree. ICCV 2011 (Oral)


Video from a Single Exposure Coded Photograph using a Learned Over-Complete Dictionary[project]

with Yas, Mohit, Tomoo, and Shree. ICCV 2011


Coded Rolling Shutter Photography: Flexible Space-Time Sampling [project]

with Yas, Tomoo, and Shree. ICCP 2010


Measurement, Modeling, and Synthesis of Time-Varying Appearance of Natural Phenomena [PDF]

Jinwei Gu. PhD Thesis. Columbia University. March 2010


Removing Image Artifacts Due to Dirty Camera Lenses and Thin Occluders[project]

with Ravi, Peter, and Shree. SIGGRAPH Asia 2009


Printing Spatially-Varying Reflectance [project]

with Wojciech Matusik, Boris Ajdin, Jason Lawrence, Hendrik P. A. Lensch, Fabio Pellacini, and Szymon Rusinkiewicz. SIGGRAPH Asia 2009


Compressive Structured Light for Recovering Inhomogenous Participating Media [project](with code)

with Ravi, Peter, Shree, and Eitan. ECCV 2008 (Oral)


Dirty Glass: Modeling and Rendering Contamination on Transparent Surfaces [project] [database] (with code)

with Ravi, Peter, and Shree. EGSR 2007


Time-Varying Surface Appearance [project][database] (with code)

with Chien-i Tu, Ravi, Peter, Wojciech, and Shree. SIGGRAPH 2006


Radiometric Calibration from a Single Image [project]

with Stephen Lin, Shuntaro Yamazaki, and Harry Shum, CVPR 2004 (Oral)


Fingerprint Orientation Field Modeling and Its Applications [project]

with Jie Zhou, Chunyu Yang. CVPR 2003


Selected Publications

Computer Vision and Machine Learning

Computational Photography

Computer Graphics and Physics-based Computer Vision

Image Processing and Biometrics

Other Areas


A Short Bio

I am a senior research scientist in NVIDIA Research. My current research focuses on deep learning for low-level and middle-level computer vision, visual SLAM, 3D reconstruction, and multi-camera systems. From August 2010 to June 2013, I was an assistant professor at the Munsell Color Science Laboratory in Center for Imaging Science in Rochester Institute of Technology. From November 2013 to October 2015, I was a senior researcher in the media lab (New Jersey Office) at Futurewei Technologies (aka, Huawei R&D USA) in Bridgewater, NJ. I obtained a Ph.D. in Spring 2010 in the Computer Science Department of Columbia University with Prof. Shree Nayar, Prof. Peter Belhumeur, and Prof. Ravi Ramamoorthi, and I got a B.S. and M.S. (supervised by Prof. Jie Zhou) in the Department of Automation in Tsinghua University, China, in 2002 and 2005 respectively. My general research interests are computer vision, machine learning, and computational imaging. Here is a copy of my CV.


Valid HTML 4.01 Transitional