I am interested in the interface between machine learning and computer vision. Areas I have worked on include multiple instance learning, object detection and localization, multiple kernel learning, structured output regression, clustering, and taxonomy discovery.

Some source code is available here.

• February 2009 - Christoph Lampert and I will be presenting a short course on kernel methods in computer vision at CVPR 2009.
• October 2008 - Learning to Localize Objects with Structured Output Regression won the best student paper prize at ECCV 2008.
• June 2008 - Beyond Sliding Windows: Object Localization by Efficient Subwindow Search won the best paper prize at CVPR 2008.
• June 2008 - Christoph Lampert and I were awarded the DAGM Main Prize for our paper A Multiple Kernel Learning Approach to Joint Multi-Class Object Detection.