Connected Component Model for Multi-Object Tracking

Abstract of Connected Multi-Object Tracking Model On MAT LAB System

Connected Component Model for Multi-Object Tracking.MULTI-OBJECT tracking aims to automatically find the moving paths of all the targets in a dynamic scene .

extensively requested by many real-world computer vision systems for intelligent video surveillance, driving assistance and intelligent robots.

A multi-object tracker must have the capability to start a path, maintain the tracking and terminate it, and thus many challenging factors seriously degrade the performance of a multi-object tracker, such as the uncertainty of the number of targets, the potential occlusion of objects during the tracking, and the irregularityof each target’s movement. Some classification based trackers need robust classifiers  to deal with the variation noise .

it is hard to balance the efficiency and the robustness of these trackers.In recent years, the rapid development of detectors  has made it possible to design an automatic multi-object tracker, evidenced by the successful practical application of ‘ trackingby-detection’  for simultaneously tracking multiple objects.

The tracking-by-detection framework consists of two parts: detection and t