Online and Real-Time Tracking in a Surveillance Scenario
This work addresses the need for efficient and accurate tracking in 24/7 surveillance systems, though it is incremental as it builds on existing real-time tracking methods.
The paper tackles the problem of real-time multiple object tracking in surveillance scenarios with static cameras, achieving state-of-the-art performance on the MOT20 benchmark by outperforming all other real-time methods in HOTA, MOTA, and IDF1 metrics.
This paper presents an approach for tracking in a surveillance scenario. Typical aspects for this scenario are a 24/7 operation with a static camera mounted above the height of a human with many objects or people. The Multiple Object Tracking Benchmark 20 (MOT20) reflects this scenario best. We can show that our approach is real-time capable on this benchmark and outperforms all other real-time capable approaches in HOTA, MOTA, and IDF1. We achieve this by contributing a fast Siamese network reformulated for linear runtime (instead of quadratic) to generate fingerprints from detections. Thus, it is possible to associate the detections to Kalman filters based on multiple tracking specific ratings: Cosine similarity of fingerprints, Intersection over Union, and pixel distance ratio in the image.