Recent Advances in Motion Understanding

Probably the most ambitious goal of Computer Vision is to build the universal vision machine, capable of guiding itself through arbitrary environments, recognizing objects along its path and reaching its destination, wherever that might be. In spite of this elusive goal, Computer Vision is a eld of research where enormous progress has been accomplished. In particular, the paradigm of Active Vision enables the research community to better address the questions related to vision applications such as autonomous navigation, visual attention, foveal and peripheral vision and real-time issues by embedding the properties of a visual system into sets of tasks to be performed and with extended control over viewing parameters, thus greatly reducing the inherent complexities of constructing a vision machine worth bearing the name. In this keynote address, we investigate the general role of motion within the paradigm of Active Vision. In particular, we examine recent results concerning motion and visual attention, motion and autonomous navigation, motion in foveal and peripheral visual areas, both in measurement and interpretation. We conclude by outlining current research areas and promising directions in Active Vision.