Binocular Tracking Based on Virtual

This paper presents a stereo active vision system which performs tracking tasks on smoothly moving objects in complex backgrounds. Dynamic control of the vergence angle adapts the horopter geometry to the target position and allows to pick it up easily on the basis of stereoscopic disparity features. We introduce a novel vergence control strategy based on the computation of \virtual horopters" to track a target movement generating rapid changes of disparity. The control strategy is implemented on a binocular head, whose right and left pan angles are controlled independently. Experimental results of gaze holding on a smoothly moving target translating and rotating in a complex surrounding demonstrate the eciency of the tracking system. 1 Introduction The importance of eye movement to biological visual systems is obvious. In contrast, controlled camera movement have played a small role in computer vision research, but are becoming increasingly recognized as important capabilities in robotic visual perception [1, 2, 3, 4]. In fact, active motion of camera provide many advantages: Since tracking involves that the visual target remains near the center of the image, it allows the use of localized visual processing and stereo algorithms that accept only a limited range of disparity. Moreover, since the eyes follow the target , the target image tends to have slow motion across the retina. On contrary, surrounding distractors move rapidly across the retina and suer from motion blur. As a result, the signal of the target is emphasized over the background. In binocular systems, whose cameras have their op-tic axis in the same plane, gaze control is the process of adjusting pan angles so that both eyes are looking at the same world point. Gaze control may be broken