Jon Barron
jonbarron@gmail.com

I am a fourth year PhD candidate in EECS at UC Berkeley. My advisor is Jitendra Malik. I am funded by the NSF GRFP.

I've spent time at Google Mountain View, MIT CSAIL, Captricity, NASA Ames Research Center, Google NYC, the NYU Media Research Lab, and the Novartis Institutes for BioMedical Research. I've worked on Astrometry.net.

I did my bachelors at the University of Toronto.

I have a CV.

Shape, Albedo, and Illumination from a Single Image of an Unknown Object
Jonathan T. Barron, Jitendra Malik
Computer Vision and Pattern Recognition (CVPR), 2012 (supplementary material) (bibtex)

We present SAIFS (shape, albedo, and illumination from shading), which produces reasonable results on arbitrary grayscale images of masked objects, and outperforms all previous grayscale "intrinsic images" algorithms.

A Category-Level 3-D Object Dataset: Putting the Kinect to Work
Allison Janoch, Sergey Karayev, Yangqing Jia, Jonathan T. Barron, Mario Fritz, Kate Saenko, Trevor Darrell
International Conference on Computer Vision (ICCV) 3DRR Workshop, 2011 (bibtex)

Using the Microsoft Kinect, we gather a large dataset of indoor crowded scenes. We investigate ways to unify state-of-the-art object detection systems and improve them with depth information.

High-Frequency Shape and Albedo from Shading using Natural Image Statistics
Jonathan T. Barron, Jitendra Malik
Computer Vision and Pattern Recognition (CVPR), 2011 (bibtex)

To solve shape-from-shading and intrinsic images simultaneously, we impose "naturalness" priors over albedo and shape and recover the most likely albedo and shape that together explain a single image.

Discovering Efficiency in Coarse-To-Fine Texture Classification
Jonathan T. Barron, Jitendra Malik
Technical Report, 2010 (bibtex)

We introduce a model and feature representation for joint texture classification and segmentation that learns how to classify accurately and when to classify efficiently. This allows for sub-linear coarse-to-fine classification.

Blind Date: Using proper motions to determine the ages of historical images
Jonathan T. Barron, David W. Hogg, Dustin Lang, Sam Roweis
The Astronomical Journal, 136, 2008

Using only raw pixel data and known catalog proper motions, it is possible to accurately estimate the date of origin of historical imagery. This allows us to retrieve lost meta-data, improve astrometric calibration, and re-estimate proper motions.