Multi‐object tracking using dominant sets. Issue 4 (19th April 2016)

• Record Type: Journal Article
• Authors: Tesfaye, Yonatan T.; Zemene, Eyasu; Pelillo, Marcello; Prati, Andrea
• Is Part Of: IET computer vision. Volume 10:Issue 4(2016)
• Page Start: 289
• Page End: 298
• Publication Date: 2016-04-19
• Subjects: object tracking -- set theory -- pattern clustering -- graph theory -- video signal processing -- computer vision -- image fusion; multiobject tracking -- computer vision -- data association technique -- locally limited temporal domain -- dominant set clustering based tracker -- auxiliary edge weighted graph -- pairwise relationship -- video temporal span -- temporal sliding window technique -- tracklet merging -- long term occlusion; Computer vision -- Periodicals; Pattern recognition systems -- Periodicals; 006.37
• DOI: 10.1049/iet-cvi.2015.0297
• Languages: English
• ISSNs: 1751-9632

Abstract

Abstract : Multi‐object tracking is an interesting but challenging task in the field of computer vision. Most previous works based on data association techniques merely take into account the relationship between detection responses in a locally limited temporal domain, which makes them inherently prone to identity switches and difficulties in handling long‐term occlusions. In this study, a dominant set clustering based tracker is proposed, which formulates the tracking task as a problem of finding dominant sets in an auxiliary edge weighted graph. Unlike most techniques which are limited in temporal locality (i.e. few frames are considered), the authors utilised a pairwise relationships (in appearance and position) between different detections across the whole temporal span of the video for data association in a global manner. Meanwhile, temporal sliding window technique is utilised to find tracklets and perform further merging on them. The authors' robust tracklet merging step renders the tracker to long term occlusions with more robustness. The authors present results on three different challenging datasets (i.e. PETS2009‐S2L1, TUD‐standemitte and ETH dataset ('sunny day' sequence)), and show significant improvements compared with several state‐of‐art methods.