VISIONTRAIN: Computational and Cognitive Vision Systems

Jan 1, 2006 · 3 min read
The EU Marie Curie Actions logo.

VISIONTRAIN: Computational and Cognitive Vision Systems

VISIONTRAIN was an EU FP6 Marie Curie Research Training Network that linked 11 European laboratories to study vision from both computational and cognitive angles. The consortium set out to narrow the gap between biological vision, which remains only partially understood, and computer vision by developing formal mathematical models and validating them experimentally. Research was organized around five themes: (i) low-level vision theory and algorithms, (ii) motion interpretation from image sequences, (iii) learning and recognising shapes, objects and categories, (iv) cognitive models of the act of seeing, and (v) functional brain-imaging for modelling visual processes.

Roles

Jan 1 2006 - Apr 30 2009: Researcher | Ph.D. Student @ FRI

  • Ph.D. research on self-supervised online learning vector quantization (SSLVQ) methods for learning object affordances using a robot arm to interact with household objects while tracking visual features.

Awards

Videos

Learning basic object affordances with a robotic arm

Video: Barry Ridge’s YouTube Channel. Credit: Barry Ridge, University of Ljubljana.

Publications

Self-Supervised Cross-Modal Online Learning of Basic Object Affordances for Developmental Robotic Systems. 2010 IEEE International Conference on Robotics and Automation, 2010.
Unsupervised Learning of Basic Object Affordances from Object Properties. Proceedings of the Fourteenth Computer Vision Winter Workshop (CVWW 2009), 2009.
Towards Learning Basic Object Affordances from Object Properties. Proceedings of the Eight International Conference on Epigenetic Robotics (EpiRob 2008), 2008.
A System for Learning Basic Object Affordances Using a Self-Organizing Map. Proceedings of the First International Conference on Cognitive Systems (CogSys 2008), 2008.
Interaktiven Sistem Za Kontinuirano Učenje Vizualnih Konceptov. Proceedings of the Sixteenth Electrotechnical and Computer Science Conference (ERK 2007), 2007.
A Framework for Continuous Learning of Simple Visual Concepts. Proceedings of the Twelveth Computer Vision Winter Workshop (CVWW 2007), 2007.
A System for Continuous Learning of Visual Concepts. Proceedings of the Fifth International Conference on Computer Vision Systems (ICVS 2007), 2007.
On Different Modes of Continuous Learning of Visual Properties. Proceedings of the Fifteenth Electrotechnical and Computer Science Conference (ERK 2006), 2006.