Taking Mobile Multi-Object Tracking to the Next Level: People, Unknown Objects, and Carried Items
Dennis Mitzel and Bastian Leibe
Abstract
In this paper, we aim to take mobile multi-object tracking to the next level by proposing a novel tracking approach that can track both known and unknown object categories in very challenging street scenes. The large variety of object types that can be encountered in such scenarios makes it infeasible to rely on pre-trained detectors. We therefore present a tracking-before-detection approach that relies on noisy stereo depth data in order to segment and track objects in 3D. At the core of our approach is a novel, compact 3D representation that allows us to robustly track a large variety of objects, to build up models of their shape online, and to use those models to detect anomalous shapes, such as carried items on a person's body. We evaluate our approach on several challenging video sequences of busy pedestrian zones and achieve state-of-the-art performance.
Results
coming soon ...
Datasets
coming soon ...