Danijel Skočaj is a researcher with the Faculty of Computer and Information Science, University of Ljubljana. He received the PhD degree from the same institution in 2003. His main research interests lie in the field of cognitive vision and artificial cognitive systems and include automatic modeling of objects from visual information with the emphasis on the robust and incremental appearance-based visual learning and recognition as well as the design and development of cognitive system with the emphasis on the interactive learning of categorical knowledge.
Activities and Events
- Workshops chair at ECCV 2006, Graz, Austria, 7-13 May 2006.
- President of SPRS, the Slovenian Pattern Recognition Society.
- Co-organizer of Cognitive Systems Kickoff, Bled, Slovenia, 28-30 October 2004.
- Program chair of Computer Vision Winter Workshop 2004, Piran, Slovenia, 2-4 February 2004.
Awards
-
Best paper award on 26th Workshop of the Austrian Association for Pattern Recognition (ÖAGM/AAPR), Graz, Austria, 10 - 11 Sept. 2002.
-
Best PhD-based paper award on Eleventh International Electrotechnical and Computer Science Conference, Portorož, Slovenia, 23 - 25 Sept. 2002.
Projects
Current Projects
- Computer Vision, Research Programme, P2-0214, 2009−2014
- Learning, analysis, and detection of motion in the framework of a hierarchical compositional visual architecture, Basic Research and Application Project, J2-4284, 2011−2014
- CogX - Cognitive Systems that Self-Understand and Self-Extend, European Project (Framework Programmes), ICT – 215181 – CogX, 2008−2012
Closed Projects
- VISIONTRAIN, Computational and Cognitive Vision Systems: A Training European Network, European Project (Framework Programmes), MRTN-CT-2004-005439, 2005−2009
- Computer Vision, Research Programme, P2-0214, 2004−2008
- Cognitive Systems for Cognitive Assistants - CoSy, European Project (Framework Programmes), FP6-004250-IP, 2004−2008
Selected publications
-
SKOČAJ, Danijel, LEONARDIS, Aleš. Incremental and robust learning of subspace representations. Image and Vision Computing, 2006. In press.
-
FIDLER, Sanja, SKOČAJ, Danijel, LEONARDIS, Aleš. Combining reconstructive and discriminative subspace methods for robust classification and regression by subsampling. IEEE trans. pattern anal. mach. intell., Mar. 2006, vol. 28, no. 3, str. 337-350.
-
SKOČAJ, Danijel, LEONARDIS, Aleš. Weighted and robust incremental method for subspace learning. V: Ninth IEEE international conference on computer vision, 13-16 October 2003,
-
SKOČAJ, Danijel, BISCHOF, Horst, LEONARDIS, Aleš. A robust PCA algorithm for building representations from panoramic images. V: HEYDEN, Anders (
Graduate Students
Barry Ridge
