The talk first presents a new type of motion feature, the co-occurrence flow (CoF), which we designed for pedestrian detection. While recent work has established a combination of oriented gradients and optic flow as effective features, our advance is to capture relative movements of different parts of the entire body, unlike existing motion features which extract internal motion in a local fashion. Through evaluations on the TUD-Brussels pedestrian dataset, we show that our motion feature based on co-occurrence flow contributes to boost the performance of existing methods.
Another topic will be about detecting bipedal motion in video sequences by using point trajectories. Given a number of point trajectories, we find a subset of points which are arising from feet in bipedal motion by analysing their spatio-temporal correlation in a pairwise fashion. To this end, we introduce probabilistic trajectories as our features which associate each point over a sufficiently long time period in the presence of noise. They are extracted from directed acyclic graphs whose edges represent temporal point correspondences and are weighted with their matching probability in terms of appearance and location. We then learn the correlation between the motion of two feet using the probabilistic trajectories in a framework of classification.
