Cognitive Systems for Cognitive Assistants (CoSy) is an EU FP6 integrated project, which aims to advance the science of cognitive systems through a multi-disciplinary investigation of architectures, representations, perceptual mechanisms, learning, planning, reasoning, action, and communication, as requirements for a human-like, autonomous, integrated, physical robot system. The Visual Cognitive Systems Laboratory VICOS participates in the project by conducting research on cognitive vision. We mainly focus on incremental visual learning of hierarchical representations and on building cognitive maps of the environment. In our research work we use mobile platforms, which are equipped with stereo camera systems, panoramic cameras, and range sensors. The main achievement of our group in the first year of the project is publication of a paper in the prestigious journal IEEE Transactions on Pattern Analysis and Machine Intelligence, in which we propose a new approach for combining reconstructive and discriminative methods for visual learning and classification.
References
- Fidler S, Skočaj D, Leonardis A. Combining Reconstructive and Discriminative Subspace Methods for Robust Classification and Regression by Subsampling. IEEE Transactions on Pattern Analysis and Machine Intelligence 28(3): 337-350, 2006. [IEEE]