Who am I?

I am graduate of the Vision Lab at UCLA. My research interests include category recognition, object class localization, and robotics. In the past, I've built large underwater robots and participated in the DARPA Grand Challenge with the Golem Group.

You can reach me via e-mail by placing bfulkers before an @ sign, followed by cs.ucla.edu.

Brian Fulkerson

Software - VLFeat - New Version Available

The newest version of VLFeat now includes an implementation of the fast dense SIFT descriptors and fast Agglomerative Information Bottleneck (AIB) described in our ECCV 2008 paper. More details may be found at the project site.

Software - Blocks

Blocks provides a matlab framework for experiments which can be described with block diagrams. It also contains an implementation of our ECCV 2008 and our ICCV 2009 papers. It requires VLFeat and libsvm-mat. This implementation also includes a baseline bag of features categorizer. More details may be found at the project site.

Software - Really quick shift: GPU segmentation - Now Available

We provide a CUDA 3.0 GPU implementation of quick shift. The implementation is very fast on modern GPUs (up to 5-10 frames per second for normal sized images). Compilation instructions and code available here.

Software - GCMex

GCMex is a MATLAB wrapper around Olga Veksler's GCoptimization toolbox (Version 2.3). Our wrapper is pre-compiled for most architectures and simple to use. More details here

Software - Patch-Based Texture Edges - MATLAB

I've implemented the texture edge detection method described in:

Patch-based Texture Edges and Segmentation, Lior Wolf, Xiaolei Huang, Ian Martin, and Dimitris Metaxas, Proc. of the 9th European Conf. on Computer vision, 2006.

Download the function here.


Finding Categories B. Fulkerson. Ph.D Thesis, University of California Los Angeles, September 2010.

VLFeat - An Open and Portable Library of Computer Vision Algorithms, A. Vedaldi and B. Fulkerson. In Proceedings of the 18th annual ACM international conference on Multimedia, October 2010. Code available.

Really quick shift: Image segmentation on a GPU, B. Fulkerson and S. Soatto. In Proceedings of the Workshop on Computer Vision using GPUs, held with the European Conference on Computer Vision, September 2010. Code available.

Class Segmentation and Object Localization with Superpixel Neighborhoods, B. Fulkerson, A. Vedaldi, and S. Soatto. In Proceedings of the International Conference on Computer Vision, 2009. Poster. Code available in Blocks.

Localizing Objects With Smart Dictionaries, B. Fulkerson, A. Vedaldi, and S. Soatto. In Proceedings of the European Conference on Computer Vision, 2008. Code available in Blocks.

Soil Sensor Technology: Life within a Pixel, M. Allen, R. Vargas, E. Graham, W. Swenson, M. Hamilton, M. Taggart, T. Harmon, A. Rat'Ko, P. Rundel, B. Fulkerson, and D. Estrin, BioScience, Vol. 57, No. 10, November 2007

Sensor Internet Share and Search: Enabling Collaboration of Citizen Scientists, S. Reddy, G. Chen, B. Fulkerson, S.J. Kim, U. Park, N. Yau, J. Cho, M. Hansen, and J. Heidemann, IPSN DSI '07 Cambridge, Massachusetts, USA

A Scalable Architecture for Multi Agent Vision Based Robot Scavenging, K. Wnuk, B. Fulkerson, and J. Sudol, Proceedings of the National Conference on Artificial Intelligence (AAAI 06) Workshop on Robotics, Boston, MA, 2006.

Autonomous off road driving in the DARPA Grand Challenge, E. Jones, B. Fulkerson, E. Frazzoli, D. Kumar, R. Walters, J. Radford, and R. Mason. In Proceedings of the IEE/ION Position, Location, and Navigation Symposium, April 2006

The Golem Group / UCLA autonomous ground vehicle in the DARPA Grand Challenge, R. Mason, J. Radford, D. Kumar, R. Walters, B. Fulkerson, E. Jones, D. Caldwell, J. Meltzer, Y. Alon, A. Shashua, H. Hattori, N. Takeda, E. Frazzoli, and S. Soatto. Journal of Field Robotics, Special Issue on the DARPA Grand Challenge 2005.