|
Andrea Vedaldi, Ph.D.
(vedaldi@robots.ox.ac.uk)
|
- 6/11/2010
- CVPR10 Tutorial on Open Source Vision Software.
- 6/11/2010
- New contributed Python interface to siftpp.
- 5/24/2010
- VLFeat 0.9.8 relased.
- 8/25/2008
- I have joined Oxford VGG as postdoctoral researcher.
- 6/21/2008
- My Ph.D. thesis is available here.
- 1/11/2007
- VicinalBoost code is now available.
Research interests: Object and category classification and detection. Invarian visual features. Structured output learning.
CVPR 2010 Tutorial on Open Source Vision Software. The Open Source Vision Software, Intro and Training CVPR 2010 tutorial aims at introducing computer vision students and researchers to a selection of open source computer vision packages. The emphasis is on problem solving, with demos illustrating how to apply the software to relevant computer vision problems. It will be discussed how the open source approach fosters learning, reproducible research, and scientific collaboration.
VLFeat. The VLFeat open source library implements popular computer vision algorithms in a simple-to-use package with MATLAB bindings. It bundles algorithms such as SIFT, MSER, k-means, hierarchical k-means, kd-trees, agglomerative information bottleneck, quick shift.
CVPR 2010 Poster
Paper
Efficient additive kernels: The homogeneous kernel map. We introduce closed-form finite dimensional feature maps approximating the additive kernels (intersection, Hellinger’s, χ2, Jensen-Shannon, ...). By adding onle line to your code you can use non-linear additive kernels as if they were linear, with vastly improved training and testing speed and compactness of the resulting models (code).