Qualitative Reasoning Feeding Back into Quantitative Model-Based Tracking

Abstract. Tracking vehicles in image sequences of innercity road traffic scenes must be considered still to constitute a challenging task. Even if a-priori knowledge about the 3D shape of vehicles, of the background structure, and about vehicle motion is provided, (partial) occlusion and dense vehicle queues easily can cause initialization and tracking failures. A stepwise improvement of the tracking approach requires numerous and time-consuming experiments. These difficulties can be eased considerably by endowing the system with – at least part of the – qualitative knowledge which a human observer activates in order to judge the results. In the case to be reported here, a system for qualitative reasoning has been coupled with a quantitative model-based tracking system in order to explore the feedback from qualitative reasoning into the geometric tracking subsystem. The approach and encouraging experimental results obtained for real-world image sequences are described.