MOVING OBJECT TRACKING BASED ON REGION LABEL ASSIGNMENT

摘要

For tracking moving objects, the template matching method is usually used. However, it cannot cope with apparent changes of a target object in images sufficiently. On the other hand, to discriminate multiple objects in still images, the label assignment based on the MAP estimation using object's features is convenient, hi this study, we propose a method which enables to track moving objects stably without explicit tracking by extending the above MAP assignment in the temporal direction. We propose two techniques; information of target position and size detected in the previous frame is propagated to the current frame as a prior probability of the target region, and distribution properties of target's feature values in a feature space are adaptively updated based on detection results at each frame. Since the proposed method is based on a label assignment and then, it is not an explicit tracking based on target appearance in images, the method is robust for occlusion. Our method can adopt various kinds of features, and in this study, we construct the algorithm using color and motion features which are often used for region detection in still images.