The purpose of this project is to develop a real-time surveillance system, where a camera could be guided remotely, by the point of gaze of a human observer, while a smart compression algorithm will be applied to the attended video stream to address the available transmission bandwidth.

A human observer will attend a location on a monitor fed by a remote camera. By capturing the human's point of gaze using an eye-tracking device, the remote camera could be manipulated in order to turn towards the attended location of the scene. In addition to that a clever compression algorithm will be used on the transmitted video stream so that most of the communication bandwidth will be allocated to high-fidelity transmission of a small region around the viewer's current point of regard, while peripheral image regions will be highly degraded and transmitted over little remaining bandwidth.

• Koch, C., Ullman, S., "Selecting one among the many: a simple network implementing shifts in selective visual attention", MIT AI Lab Memo 770, January 1984.
• E. Niebur, L. Itti, C. Koch, "Controlling the Focus of Visual Selective Attention", Models of Neural Networks IV, (L. Van Hemmen, E. Domany, J. Cowan Ed.), Springer Verlag, Aug 2001.
• Backer, G., Mertsching, B., Bollmann, M., "Data- and model-driven gaze control for an active-vision system", IEEE Transactions on Pattern Analysis and Machine Intelligence, Volume: 23, Issue: 12, Dec 2001.
• Dayan, P., Kakade, S., Montague, "Learning and Selective Attention", Nature Neuroscience 3 PR 2000.
• Tsotsos, J.K., Pomplun, M., Martinez-Trujillo, J.C. & Zhou, K., "Attending to Visual Motion: Localizing and Classifying Affine Motion Patterns", Proceedings of the Canadian Conference on Computer and Robot Vision, May 17 - 20, 2004, London, Ontario.
• Itti L., "Automatic Foveation for Video Compression Using a Neurobiological Model of Visual Attention", IEEE Transactions on Image Processing, Vol. 13, No. 10, pp. 1304-1318, Oct 2004.