About the book

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.
- Revises methods to evaluate the performance of video trackers – this is identified as a major problem by end-users.
- Includes a companion website hosting slides supporting the book material and a collection of software resources and publicly available datasets to help the reader develop and test a video-based tracker


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. 


Chapter 1 introduces the video-tracking problem and presents it in a unified view by dividing the problem into five main logical tasks. Next, the chapter provides a formal problem formulation for video tracking. Finally, it discusses typical challenges that make video tracking difficult. Chapter 2 discusses current and emerging applications of video tracking. Application areas include media production, medical data processing, surveillance, business intelligence, robotics, telecollaboration, interactive gaming and art.


Chapter 3 offers a high-level overview of the video acquisition process and presents relevant features that can be selected for the representation of a target. Next, Chapter 4 discusses various shape-approximation strategies and appearance modelling techniques.


Chapter 5 introduces a taxonomy for localisation algorithms and compares single and multi-hypothesis strategies. Chapter 6 discusses the modalities for fusing multiple features for target tracking. Advantages and disadvantages of fusion at the tracker level and at the feature level are discussed. Moreover, we present appropriate measures for quantifying the reliability of features prior to their combination.


Chapter 7 extends the concepts covered in the first part of the book to tracking a variable number of objects. To better exemplify these methods, particular attention is given to multi-hypothesis data-association algorithms applied to video surveillance. Moreover, the chapter discusses and evaluates the first video-based multi-target tracker based on finite set statistics. Using this tracker as an example, Chapter 8 discusses how modelling the scene can help improve the performance of a video tracker. In particular, we discuss automatic and interactive strategies for learning areas of interest in the image. Chapter 9 covers protocols to be used to formally evaluate a video tracker and the results it generates. The chapter provides the reader with a comprehensive overview of performance measures and a range of evaluation datasets.


Finally, the Epilogue summarises the current directions and future challenges in video tracking and offers a list of further readings, while the Appendix reports and discusses comparative numerical results of selected methods presented in the book.