Video Tracking

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

Video Tracking provides a comprehensive treatment of thefundamental aspects of algorithm and application development forthe task of estimating, over time, the position of objects ofinterest seen through cameras. Starting from the general problemdefinition and a review of existing and emerging video trackingapplications, the book discusses popular methods, such as thosebased on correlation and gradient-descent. Using practicalexamples, the reader is introduced to the advantages andlimitations of deterministic approaches, and is then guided towardmore advanced video tracking solutions, such as those based on theBayes' recursive framework and on Random Finite Sets.Key features:* Discusses the design choices and implementation issues requiredto turn the underlying mathematical models into a real-worldeffective tracking systems.* Provides block diagrams and simil-code implementation of thealgorithms.* 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 developtechniques and solutions based on the potential of video trackingapplications. The design methodologies discussed throughout thebook provide guidelines for developers in the industry working onvision-based applications. The book may also serve as a referencefor engineering and computer science graduate students involved invision, robotics, human-computer interaction, smart environmentsand virtual reality programmes

Foreword.About the authors.Preface.Acknowledgments.Notations.Acronyms.1 What is video tracking?1.1 Introduction.1.2 The design of a video tracker.1.3 Problem formulation.1.4 Interactive versus automated tracking.1.5 Summary.2 Applications.2.1 Introduction.2.2 Media production and augmented reality.2.3 Medical applications and biological research.2.4 Surveillance and business intelligence.2.5 Robotics and unmanned vehicles.2.6 Tele-collaboration and interactive gaming.2.7 Art installations and performances.2.8 Summary.References.3 Feature extraction.3.1 Introduction.3.2 From light to useful information.3.3 Low-level features.3.4 Mid-level features.3.5 High-level features.3.6 Summary.References.4 Target representation.4.1 Introduction.4.2 Shape representation.4.3 Appearance representation.4.4 Summary.References5 Localisation.5.1 Introduction.5.2 Single-hypothesis methods.5.3 Multi-hypothesis methods.5.4 Summary.References.6 Fusion.6.1 Introduction.6.2 Fusion strategies.6.3 Feature fusion in a Particle Filter.6.4 Summary.References.7 Multi-target management.7.1 Introduction.7.2 Measurement validation.7.3 Data association.7.4 Random Finite Sets for tracking.7.5 Probabilistic Hypothesis Density filter.7.6 The Particle PHD filter.7.7 Summary.References.8 Context modeling.8.1 Introduction.8.2 Tracking with context modelling.8.3 Birth and clutter intensity estimation.8.4 Summary.References.9 Performance evaluation.9.1 Introduction.9.2 Analytical vs. empirical methods.9.3 Ground truth.9.4 Evaluation scores.9.5 Comparing trackers.9.6 Evaluation protocols.9.7 Datasets.9.8 Summary.References.Epilogue.Further reading.Appendix A: Comparative results.A.1 Single versus structural histogram.A.1.1 Experimental setup.A.1.2 Discussion.A.2 Localisation algorithms.A.2.1 Experimental setup.A.2.2 Discussion.A.3 Multi-feature fusion.A.3.1 Experimental setup.A.3.2 Reliability scores.A.3.3 Adaptive versus non-adaptive tracker.A.3.4 Computational complexity.A.4 PHD filter.A.4.1 Experimental setup.A.4.2 Discussion.A.4.3 Failure modalities.A.4.4 Computational cost.A.5 Context modelling.A.5.1 Experimental setup.A.5.2 Discussion.References.Index.