Abstract-In this article, we propose a visual pedestrian detection system which couples pedestrian appearance and pedestrian motion in a Bayesian fashion, with the goal of making detection more invariant to appearance changes. More precisely, the system couples dense appearance-based pedestrian likelihoods derived from a sliding-window SVM detector to spatial prior distributions obtained from the prediction step of a particle filter based pedestrian tracker. This mechanism, which we term dynamic attention priors (DAP), is inspired by recent results on predictive visual attention in humans and can be implemented at negligible computational cost. We prove experimentally, using a set of public, annotated pedestrian sequences, that detection performance is improved significantly, especially in cases where pedestrians differ from the learned models, e.g., when they are too small, have an unusual pose or occur before strongly structured backgrounds. In particular, dynamic attention priors allow to use more restrictive detection thresholds without losing detections while minimizing false detections.

In this contribution, we propose to use road and lane information as contextual cues in order to increase the precision of multi-object object tracking. For tracking, we employ a Monte Carlo implementation of a Probability Hypothesis Density (PHD)-filter, whereas scene context (road and lane information) is taken from annotated street maps. The novel aspect of the presented work is the tightly coupling of context information and the particle filtering process. This is achieved by injecting a priori particles representing locally expected motions, which are in turn determined by the local road and the lane configuration. This approach is evaluated on objects (tracklets) from the public KITTI benchmark database. Our experimental findings demonstrate a considerable tracking precision increasing when including this kind of a priori knowledge. At the same time, the approach is able to determine objects whose movements differ from the locally expected motion, which is an important feature for safety applications.