Object Tracking Using Mean Shift and Kalman Filter: APerformance Comparison on Public Datasets

Abstract

Object tracking in video remains a fundamental computer vision challenge,especially when faced with real-world complexities like occlusion and dynamic motion. Thisreview offers a comparative analysis of two enduring methodologies, the Mean Shift algo-rithm and the Kalman Filter, focusing on research published between 2020 and 2025. MeanShift, a non-parametric tracker, relies on appearance features, while the Kalman Filter, astate estimator, models object motion. We synthesize recent findings on their design, per-formance, and limitations, drawing on evaluations from standard benchmarks like MOT,KITTI, and PETS. The analysis highlights a strong trend towards hybrid approaches thatleverage the complementary strengths of these classical techniques to achieve robust, real-time tracking in demanding scenarios.