3D tracking based on possibilities rather than probabilities Application to flying honeybees at the beehive entrance

This article covers a tracking appoach for multiple simultaneously flying animals. The idea is to constraint the tracking to the physically possible space without any knowledge of the scene. Accordingly, we propose to use a naively made trajectory dataset as a base to track the targets into the possible space rather than the probable space. First, a trajectory database is computed under weak assumptions (permissive classic Kalman Filter). Then, for the effective tracking, we replace the prediction phase of the Kalman Filter to rely only on the dynamic stored in the database of previous trajectories. We show that performances can reach these obtained by other methods based on strong assumptions (e.g. the 3D model of the scene).