Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking

🤖 Plain-English Summary

Kalman filter (KF) based methods for multi-object tracking (MOT) make an assumption that objects move linearly. It achieves advanced on multiple datasets, including MOT17, MOT20, KITTI, head tracking, and especially DanceTrack where the object motion is highly non-linear.

🔑 Key Findings

• While this assumption is acceptable for very short periods of occlusion, linear estimates of motion for prolonged time can be highly inaccurate.
• Moreover, when there is no measurement available to update Kalman filter parameters, the standard convention is to trust the priori state estimations for posteriori update.
• This leads to the accumulation of errors during a period of occlusion.

🔗 Related Resources