Record:   Prev Next
Author Betke, Margrit, author
Title Data association for multi-object visual tracking / Margrit Betke, Zheng Wu
Imprint [San Rafael, California] : Morgan & Claypool, 2017
Descript 1 online resource (ix, 110 pages) : illustrations
still image rdacontent
text rdacontent
electronic isbdmedia
online resource rdacarrier
Series Synthesis lectures on computer vision, 2153-1064 ; # 9
Synthesis digital library of engineering and computer science
Synthesis lectures on computer vision ; # 9. 2153-1064
Note Part of: Synthesis digital library of engineering and computer science
Includes bibliographical references (pages 85-108)
8. Application to animal group tracking in 3D: 8.1. Two sample systems for analyzing bat and bird flight; 8.2. Impact of multi-animal tracking systems -- 9. Benchmarks for human tracking: 9.1. PETS-2009; 9.2. Beyond PETS-2009: the MOT-challenge benchmark -- 10. Concluding remarks -- Bibliography -- Authors' biographies
Preface -- 1. An introduction to data association in computer vision: 1.1. Challenges; 1.2. Related topics beyond the scope of this book; 1.3. Application domains; 1.4. Simulation testbeds; 1.5. Experimental benchmarks; 1.6. Organization of the book -- 2. Classic sequential data association approaches: 2.1. Advantages of Kalman filters for use in multi-object tracking; 2.2. Gating; 2.3. Global nearest neighbor standard filter (GNNSF); 2.4. Joint probabilistic data association (JPDA); 2.5. Multiple hypotheses tracking (MHT); 2.6. Discussion -- 3. Classic batch data association approaches: 3.1. Markov chain Monte Carlo data association (MCMCDA); 3.2. Network flow data association (NFDA); 3.3. Probabilistic multiple hypothesis tracking (PMHT); 3.4. Discussion -- 4. Evaluation criteria: 4.1. Definitions; 4.2. Discussion -- 5. Tracking with multiple cameras: 5.1. The reconstruction-tracking approach; 5.2. The tracking-reconstruction approach; 5.3. An example of spatial data association; 5.4. Discussion -- 6. The tracklet linking approach: 6.1. Review of existing work; 6.2. An example of tracklet linking using a track graph -- 7. Advanced techniques for data association: 7.1. Data association for merged or split measurements; 7.2. Learning-based data association; 7.3. Coupling data association --
Abstract freely available; full-text restricted to subscribers or individual document purchasers
Google scholar
Google book search
System requirements: Adobe Acrobat Reader
Mode of access: World Wide Web
This book serves as a tutorial on data association methods, intended for both students and experts in computer vision. We describe the basic research problems, review the current state of the art, and present some recently developed approaches. The book covers multi-object tracking in two and three dimensions. We consider two imaging scenarios involving either single cameras or multiple cameras with overlapping fields of view, and requiring across-time and across-view data association methods. In addition to methods that match new measurements to already established tracks, we describe methods that match trajectory segments, also called tracklets. The book presents a principled application of data association to solve two interesting tasks: first, analyzing the movements of groups of free-flying animals and second, reconstructing the movements of groups of pedestrians. We conclude by discussing exciting directions for future research
Also available in print
Title from PDF title page (viewed on October 21, 2016)
Link Print version: 9781627059558
Subject Data integration (Computer science)
Computer vision -- Mathematical models
Automatic tracking -- Mathematical models
multi-target tracking
multi-object tracking
data association
multi-view tracking
multi-camera tracking
tracklet association
tracklet linking
tracklet stitching
tracking evaluation
MOT evaluation
Bayesian recursive filter
Bayesian multi-target tracking
Bayesian multi-object tracking
people tracking
animal tracking
group tracking
tracking bats
tracking birds
Alt Author Wu, Zheng, author
Record:   Prev Next