Simple online and real-time tracking with occlusion handling
This work addresses the challenge of real-time tracking for applications like surveillance or autonomous driving by improving accuracy without sacrificing speed, though it is incremental as it builds on existing motion-based approaches.
The paper tackles the problem of online multiple object tracking by proposing an algorithm that uses only geometric cues to handle occlusions and re-identification, resulting in a 40% reduction in identity switch and a 28% reduction in fragmentation compared to state-of-the-art online methods.
Multiple object tracking is a challenging problem in computer vision due to difficulty in dealing with motion prediction, occlusion handling, and object re-identification. Many recent algorithms use motion and appearance cues to overcome these challenges. But using appearance cues increases the computation cost notably and therefore the speed of the algorithm decreases significantly which makes them inappropriate for online applications. In contrast, there are algorithms that only use motion cues to increase speed, especially for online applications. But these algorithms cannot handle occlusions and re-identify lost objects. In this paper, a novel online multiple object tracking algorithm is presented that only uses geometric cues of objects to tackle the occlusion and reidentification challenges simultaneously. As a result, it decreases the identity switch and fragmentation metrics. Experimental results show that the proposed algorithm could decrease identity switch by 40% and fragmentation by 28% compared to the state of the art online tracking algorithms. The code is also publicly available.