We propose a real-time multiobject-tracking approach that is minimally affected by environmental conditions and target appearance change. The aim of the proposed approach is to track any object in a scene, regardless of object type, since tracking all of the objects in a scene is critical and widely used in surveillance applications. Thus, motion detection results are used to initialize the trackers. The proposed object-tracking approach is realized with two types of independent correlation filters estimating location and scale. Alternative correlation filters representing different appearances of the target are also proposed in order to increase the robustness of the approach to scene and target changes. Tracking sustainability is provided by putting alternative correlation filters into use when the quality of the tracking output decreases to a critical level. Motion blobs are also used to minimize object boundary drift, which is a challenging problem, especially for long-term tracking. The proposed approach was tested on an object-tracking benchmark dataset and outperformed most state-of-the-art methods.
Correlation filter, visual object tracking, real-time video processing
BAŞKURT, KEMAL BATUHAN and SAMET, REFİK
"Long-term multiobject tracking using alternative correlation filters,"
Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 26:
5, Article 7.
Available at: https://journals.tubitak.gov.tr/elektrik/vol26/iss5/7