These are publications I made during my PhD at ESAT-PSI/VISICS and preceding studies. All documents are in PDF format. Due to space and bandwidth constraints, I have not put the datasets online that accompany these publications. These sets include the multi-view motorbike and sports shoes sets from the CVPR06 and IJRR09 papers, and the car and wheelchair sets from the ICCV07 and CVIU09 papers. If you want to obtain those datasets you should first ask the people at VISICS (especially Tinne Tuytelaars) if they can provide them. If not, you can ask me and I'll see if I can dig them up and figure out a way to get them to you.
Mind that I do not have the rights to share the source code of the entire multi-view recognition system. If you are interested in this source code for research, you should again contact ESAT-PSI/VISICS directly.
A. Thomas, Multi-View Object Class Recognition and Metadata Transfer, PhD dissertation,
Katholieke Universiteit Leuven, ESAT-PSI, December 21st, 2009.
Promotor: Luc Van Gool
A. Thomas, V. Ferrari, B. Leibe, T. Tuytelaars, L. Van Gool, Using Multi-view Recognition and Meta-data Annotation to Guide a Robot's Attention, International Journal of Robotics Research (2009),
Vol. 28, No. 8, August 2009, pp. 976—998,
A. Thomas, V. Ferrari, B. Leibe, T. Tuytelaars, L. Van Gool, Shape-from-recognition: Recognition enables meta-data transfer, Comput. Vis. Image Understand. (2009),
Vol. 113, Issue 12, December 2009, pp. 1222—1234,
A. Thomas, V. Ferrari, B. Leibe, T. Tuytelaars, L. Van Gool, Depth-From-Recognition: Inferring Meta-data by Cognitive Feedback, Workshop on 3d Representation for Recognition, Proceedings IEEE International Conference on Computer Vision, 2007.
A. Thomas, V. Ferrari, B. Leibe, T. Tuytelaars, B. Schiele, L. Van Gool, Towards Multi-View Object Class Detection, Proceedings IEEE Conference on Computer Vision and Pattern Recognition, Vol. 2, pp. 1589-1596, 2006.
Software: the Image Exploration part used in this paper is available for download. Read the enclosed README.txt and INSTALL files for more information. Download (version 1.1, 7-zipped TAR file, 2.5MB)
Note: Somehow, a reference to this paper with incorrectly spelled names has made its way into the research community. If you encounter a citation like “A. Thomas, V. Ferrar, B. Leibe, T. Tuytelaars, B. Schiel, and L. V. Gool,” please correct the names “Ferrari” and “Schiele” if possible.
A. Thomas Kernel Methods for Improving Text Search Engines by Using One-Class Classification, Master Thesis in Master of Artificial Intelligence, Catholic University of Leuven, 2003.
A. Thomas Visual Modeling (Visueel Modelleren), Master Thesis in Electrotechnical Engineering, Catholic University of Leuven, 2002.