FollowMe: Efficient Online Min-Cost Flow Tracking with Bounded Memory and Computation

One of the most popular approaches to multi-target tracking is tracking-by-detection. Current min-cost flow algorithms which solve the data association problem optimally have three main drawbacks: they are computationally expensive, they assume that the whole video is given as a batch and they scale badly in memory and computation with the length of the video sequence. In this paper, we address each of these issues. Our first contribution is a dynamic version of the successive shortest-path algorithm which solves the data association problem optimally while reusing computation, resulting in one order of magnitude faster inference than standard solvers. With our second contribution we address the optimal solution to the data association problem when dealing with an incoming stream of data (i.e., online setting). Our last contribution is an approximate online solution with bounded memory and computation, which can handle videos of arbitrarily length while performing tracking in real time. We demonstrate the effectiveness of our algorithms on the KITTI and PETS2009 benchmarks and show state-of-the-art performance, while being significantly faster than existing solvers.