VLFeat is a cross-platform open source collection of vision algorithms with a special focus on visual features (for instance SIFT and MSER) and clustering (k-means, hierarchical k-means, agglomerative information bottleneck). It bundles a MATLAB toolbox, a clean and portable C library and a number of command line utilities. Thus it is possible to use the same algorithm both from MATLAB, the command line, and your own programs.
Many parties make the development of VLFeat possible:
Sponsors
In 2012 the development of VLFeat is
supported by the PASCAL Harvest programme. Several people have been
working in Oxford to add new functionalities to the library. Moreover,
leading researchers in computer vision were consulted as
advisors of the project.
Some of these advancements are going to be presented at an upcoming tutorial at the European Conference in Computer Vision (ECCV) 2012.
See the PASCAL Harvest Roadmap for further details.
Team
VLFeat is developed by a team of computer vision researchers, including master and PhD students, postgraduates, and senior researchers, as listed below. VLFeat was created by Andrea Vedaldi and Brian Fulkerson in 2007, based on previously published software by the same authors.
Andrea Vedaldi joined
the faculty at the University of Oxford in 2012. Since 2008 he was
junior research fellow in the Oxford Visual Geometry Group. He
received the Ph.D. and Master's degrees in Computer Science from the
University of California - Los Angeles, in 2008 and 2005, and the
Bachelor's degree in Information Engineering from the University of
Padova, Italy, in 2003. He is the recipient of the UCLA outstanding
Master's and Ph.D. awards.
Brian Fulkerson received
his B.S. in Computer Engineering from the University of California -
San Diego in 2004, and his M.S. and Ph.D. in Computer Science from
the University of California - Los Angeles Vision Lab in 2006 and
2010.
Karel Lenc Karel Lenc received his
B.S. degree in Cybernetics and Measurements from the Czech Technical
University in 2010. He is currently pursuing M.S. degree in Computer
Vision and Computer Engineering at the CTU, Center for Machine
Perception. He visited the Department of Engineering Mathematics,
University of Bristol in 2011 and the Oxford Visual Geometry Group in
2012.
Daniele Perrone
received his B.S. degree in Computer Engineering from Università
della Calabria in 2007, and his M.S. degree in Artificial
Intelligence from "Sapienza" Università di Roma in 2009. He started
his Ph.D. in June 2010 at Heriot-Watt University, Edinburgh,
UK. From June 2012 he joined the Computer Vision Group at the Universität Bern,
Switzerland.
Michal Perdoch
received his Bachelor's degree in Software Engineering from the
Slovak University of Technology in Bratislava in 2001, Master's
degree in Computer Science and Ph.D. degree in Artificial
Intelligence and Biocybernetics from the Czech Technical University
in Prague, in 2004 and 2011. He is currently a postdoctoral
researcher at the CTU Center for Machine Perception.
Advisors
The development of VLFeat is supported by a number of computer vision groups and researchers:
- Prof. Andrew Zisserman, Oxford VGG Lab.
- Prof. Jiri Matas, Czech Technical University in Prague.
- Prof. Tinne Tuytelaars, KU Leuven.
- Dr. Cordelia Schmid, LEAR, Grenoble.
- Dr. Krystian Mikolajczyk, Surrey.
- Prof. Stefano Soatto, UCLA Vision Lab.
Users and community
The authors would like to thank the many colleagues that have contributed to VLFeat by testing and providing helpful suggestions and comments.