Vol 3 , Issue 2 , April - June 2015 | Pages: 68-71 | Research Paper
Received: April 05, 2015 | Revised: April 30, 2015 | Accepted: May 20, 2015 | Published Online: June 15, 2015
Author Details
( * ) denotes Corresponding author
Multi-object tracking is still a challenging task in computer vision. A robust approach is proposed to realize multi-object tracking using camera networks. Detection algorithms are utilized to detect object regions with confidence scores for initialization of individual particle filters. Since data association is the key issue in Tracking-by-Detection mechanism, an efficient HOG algorithm and SVM classifier algorithm are used for tracking multiple objects. Furthermore, tracking in single cameras is realized by a greedy matching method. Afterwards, 3D geometry positions are obtained from the rectangular relationship between objects. Corresponding objects are tracked in cameras to take the advantages of camera based tracking. The proposed algorithm performs online and does not need any information about the scene, any restrictions of enter-and-exit zones, no assumption of areas where objects are moving on and can be extended to any class of object tracking. Experimental results show the benefits of using camera by the higher accuracy and detect the objects.
Keywords
Online Multi-Object Tracking; Tracking-By Detection; Data Association; Track Management; Online Learning; Track Existence Probability; Particle Filtering; Affinity Model; Surveillance System