![Cover image for Video tracking : theory and practice Cover image for Video tracking : theory and practice](/client/assets/5.0.0/ctx//client/images/no_image.png)
Available:*
Library | Item Barcode | Call Number | Material Type | Item Category 1 | Status |
---|---|---|---|---|---|
Searching... | 30000010273854 | TK6680.3 M34 2011 | Open Access Book | Book | Searching... |
On Order
Summary
Summary
Video Tracking provides a comprehensive treatment of the fundamental aspects of algorithm and application development for the task of estimating, over time, the position of objects of interest seen through cameras. Starting from the general problem definition and a review of existing and emerging video tracking applications, the book discusses popular methods, such as those based on correlation and gradient-descent. Using practical examples, the reader is introduced to the advantages and limitations of deterministic approaches, and is then guided toward more advanced video tracking solutions, such as those based on the Bayes' recursive framework and on Random Finite Sets.
Key features:
Discusses the design choices and implementation issues required to turn the underlying mathematical models into a real-world effective tracking systems. Provides block diagrams and simil-code implementation of the algorithms. Reviews methods to evaluate the performance of video trackers - this is identified as a major problem by end-users.The book aims to help researchers and practitioners develop techniques and solutions based on the potential of video tracking applications. The design methodologies discussed throughout the book provide guidelines for developers in the industry working on vision-based applications. The book may also serve as a reference for engineering and computer science graduate students involved in vision, robotics, human-computer interaction, smart environments and virtual reality programmes
Author Notes
Dr Emilio Maggio, Vicon, UK
Dr Maggio is Computer Vision Scientist at Vicon, the motion capture worldwide market leader. From 2004 - 2008 he was a Ph.D. student at the Department of Electronic Engineering, Queen Mary, University of London. In 2005 and again in 2007 he was awarded the best student paper prize at ICASSP. Dr Maggio has acted as a reviewer for the IEEE Transactions on Circuits and Systems for Video Technology, the International Journal of Image and Graphics and ACM Multimedia .
Dr Andrea Cavallaro, School of Electronic Engineering and Computer Science, Queen Mary, University of London, UK
Dr Cavallaro is Reader in Multimedia Signal Processing at Queen Mary, University of London. He is the author of more than 70 papers, including 5 book chapters. He is an elected member of the IEEE Signal Processing Society, Multimedia Signal Processing Committee. He has been a member of the organizing/ technical committee for several international conferences such as Technical Chair of EUSIPCO 08 and General Chair of the IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS 2007), with General Chair positions being held for forthcoming 2009 conferences such as BMVC 09. He has been guest editor of several special issues, including 'Multi-sensor object detection and tracking', Signal, Image and Video Processing (Springer). Dr Cavallaro was awarded the Royal Academy of Engineering teaching prize in 2007.
Table of Contents
Foreword | p. xi |
About the authors | p. xv |
Preface | p. xvii |
Acknowledgements | p. xix |
Notation | p. xxi |
Acronyms | p. xxiii |
1 What is video tracking? | p. 1 |
1.1 Introduction | p. 1 |
1.2 The design of a video tracker | p. 2 |
1.2.1 Challenges | p. 2 |
1.2.2 Main components | p. 6 |
1.3 Problem formulation | p. 7 |
1.3.1 Single-target tracking | p. 7 |
1.3.2 Multi-target tracking | p. 10 |
1.3.3 Definitions | p. 11 |
1.4 Interactive versus automated tracking | p. 12 |
1.5 Summary | p. 13 |
2 Applications | p. 15 |
2.1 Introduction | p. 15 |
2.2 Media production and augmented reality | p. 16 |
2.3 Medical applications and biological research | p. 17 |
2.4 Surveillance and business intelligence | p. 20 |
2.5 Robotics and unmanned vehicles | p. 21 |
2.6 Tele-collaboration and interactive gaming | p. 22 |
2.7 Art installations and performances | p. 22 |
2.8 Summary | p. 23 |
References | p. 24 |
3 Feature extraction | p. 27 |
3.1 Introduction | p. 27 |
3.2 From light to useful information | p. 28 |
3.2.1 Measuring light | p. 28 |
3.2.2 The appearance of targets | p. 30 |
3.3 Low-level features | p. 32 |
3.3.1 Colour | p. 32 |
3.3.2 Photometric colour invariants | p. 39 |
3.3.3 Gradient and derivatives | p. 42 |
3.3.4 Laplacian | p. 47 |
3.3.5 Motion | p. 49 |
3.4 Mid-level features | p. 50 |
3.4.1 Edges | p. 50 |
3.4.2 Interest points and interest regions | p. 51 |
3.4.3 Uniform regions | p. 56 |
3.5 High-level features | p. 61 |
3.5.1 Background models | p. 62 |
3.5.2 Object models | p. 63 |
3.6 Summary | p. 65 |
References | p. 65 |
4 Target representation | p. 71 |
4.1 Introduction | p. 71 |
4.2 Shape representation | p. 72 |
4.2.1 Basic models | p. 72 |
4.2.2 Articulated models | p. 73 |
4.2.3 Deformable models | p. 74 |
4.3 Appearance representation | p. 75 |
4.3.1 Template | p. 76 |
4.3.2 Histograms | p. 78 |
4.3.3 Coping with appearance changes | p. 83 |
4.4 Summary | p. 84 |
References | p. 85 |
5 Localisation | p. 89 |
5.1 Introduction | p. 89 |
5.2 Single-hypothesis methods | p. 90 |
5.2.1 Gradient-based trackers | p. 90 |
5.2.2 Bayes tracking and the Kalman filter | p. 95 |
5.3 Multiple-hypothesis methods | p. 98 |
5.3.1 Grid sampling | p. 99 |
5.3.2 Particle filter | p. 101 |
5.3.3 Hybrid methods | p. 105 |
5.4 Summary | p. 111 |
References | p. 111 |
6 Fusion | p. 115 |
6.1 Introduction | p. 115 |
6.2 Fusion strategies | p. 116 |
6.2.1 Tracker-level fusion | p. 116 |
6.2.2 Measurement-level fusion | p. 118 |
6.3 Feature fusion in a Particle Filter | p. 119 |
6.3.1 Fusion of likelihoods | p. 119 |
6.3.2 Multi-feature resampling | p. 121 |
6.3.3 Feature reliability | p. 123 |
6.3.4 Temporal smoothing | p. 126 |
6.3.5 Example | p. 126 |
6.4 Summary | p. 128 |
References | p. 128 |
7 Multi-target management | p. 131 |
7.1 Introduction | p. 131 |
7.2 Measurement validation | p. 132 |
7.3 Data association | p. 134 |
7.3.1 Nearest neighbour | p. 134 |
7.3.2 Graph matching | p. 136 |
7.3.3 Multiple-hypothesis tracking | p. 139 |
7.4 Random Finite Sets for tracking | p. 143 |
7.5 Probabilistic Hypothesis Density filter | p. 145 |
7.6 The Particle PHD filter | p. 147 |
7.6.1 Dynamic and observation models | p. 149 |
7.6.2 Birth and clutter models | p. 151 |
7.6.3 Importance sampling | p. 151 |
7.6.4 Resampling | p. 152 |
7.6.5 Particle clustering | p. 156 |
7.6.6 Examples | p. 160 |
7.7 Summary | p. 163 |
References | p. 165 |
8 Context modeling | p. 169 |
8.1 Introduction | p. 169 |
8.2 Tracking with context modelling | p. 170 |
8.2.1 Contextual information | p. 170 |
8.2.2 Influence of the context | p. 171 |
8.3 Birth and clutter intensity estimation | p. 173 |
8.3.1 Birth density | p. 173 |
8.3.2 Clutter density | p. 179 |
8.3.3 Tracking with contextual feedback | p. 181 |
8.4 Summary | p. 184 |
References | p. 184 |
9 Performance evaluation | p. 185 |
9.1 Introduction | p. 185 |
9.2 Analytical versus empirical methods | p. 186 |
9.3 Ground truth | p. 187 |
9.4 Evaluation scores | p. 190 |
9.4.1 Localisation scores | p. 190 |
9.4.2 Classification scores | p. 193 |
9.5 Comparing trackers | p. 196 |
9.5.1 Target life-span | p. 197 |
9.5.2 Statistical significance | p. 198 |
9.5.3 Repeatibility | p. 198 |
9.6 Evaluation protocols | p. 199 |
9.6.1 Low-level protocols | p. 199 |
9.6.2 High-level protocols | p. 203 |
9.7 Datasets | p. 207 |
9.7.1 Surveillance | p. 207 |
9.7.2 Human-computer interaction | p. 212 |
9.7.3 Sport analysis | p. 215 |
9.8 Summary | p. 220 |
References | p. 220 |
Epilogue | p. 223 |
Further reading | p. 225 |
Appendix A Comparative results | p. 229 |
A.1 Single versus structural histogram | p. 229 |
A.1.1 Experimental setup | p. 229 |
A.1.2 Discussion | p. 230 |
A.2 Localisation algorithms | p. 233 |
A.2.1 Experimental setup | p. 233 |
A.2.2 Discussion | p. 235 |
A.3 Multi-feature fusion | p. 238 |
A.3.1 Experimental setup | p. 238 |
A.3.2 Reliability scores | p. 240 |
A.3.3 Adaptive versus non-adaptive tracker | p. 242 |
A.3.4 Computational complexity | p. 248 |
A.4 PHD filter | p. 248 |
A.4.1 Experimental setup | p. 248 |
A.4.2 Discussion | p. 250 |
A.4.3 Failure modalities | p. 251 |
A.4.4 Computational cost | p. 255 |
A.5 Context modelling | p. 257 |
A.5.1 Experimental setup | p. 257 |
A.5.2 Discussion | p. 257 |
References | p. 261 |
Index | p. 263 |