Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Optimization for computer vision : an introduction to core concepts and methods
Treiber M., Springer Publishing Company, Incorporated, New York, NY, 2013. 257 pp. Type: Book (978-1-447152-82-8)
Date Reviewed: Apr 11 2014

The field of optimization is not unique to computer vision. Many other disciplines such as economics, operations research, and engineering utilize optimization techniques; however, optimization plays a fundamental role within computer vision, so it is useful to examine. Put simply, optimization seeks to find the “best” solution to a problem according to some criterion. The example used at the start of Treiber’s book is that of a satellite navigation system, where the user wishes to travel from place A to place B and the system searches for the optimal route based on criteria such as travel time or distance. Within computer vision, optimization appears in diverse areas such as segmentation, denoising, registration, and image reconstruction. It is the purpose of this book to introduce the reader to how optimization can be used within such areas.

The book is loosely structured around the different approaches to optimization within computer vision: continuous, discrete, combinatorial, and variational. The opening chapter seeks to quickly bring the reader up to speed with core concepts related to optimization, as well as explain the differences between each of these optimization problems. The chapter also introduces fundamental concepts such as graphs and Markov random fields. However, it is in the first chapter that we see one of the weaknesses of this book. The structure of each chapter is very abrupt; there are few or no introductory comments and there are no conclusions at the end. While such text can often be redundant and unnecessary, in the case of this book, I think it would have been really helpful to the reader. Concluding remarks would help draw each chapter to a close and would greatly improve the usefulness of the book as both a teaching and a reference tool.

That being said, the scope of material within this book is impressive. Most of the optimization approaches detailed within the book are accompanied by a technical description, a flowchart describing the structure of the algorithm, and pseudocode; most are also shown alongside an example application. For example, when the graph cuts framework is described in chapter 6, not only are the technical and algorithmic details explained, but example uses of the framework are also shown. This helps to improve the readability and applicability of this book since the content is never far away from applications and pseudocode. It is often the case with such books that the authors drift off into theory, leaving readers to try to work out how such theory could be used in practice. It is fair to say that Treiber never leaves readers in such a position; rather, he guides them through the information, from theory to application.

It is a shame that the book simply and abruptly ends. As with the chapters, the book includes no concluding remarks. It finishes at the end of chapter 7, “Dynamic Programming,” with no attempt to draw together what was described in the previous chapters. I think it would have been interesting and useful to include a final concluding chapter to point readers to other sources of information and also to discuss current problems within the optimization field. Such remarks would enable this book to be used as a springboard for readers to enter into the area of optimization research within computer vision.

Overall, this is an impressive book that will prove useful for post-graduate and advanced undergraduate readers. The constant use of pseudocode and the grounding within applications and examples make this book suitable as both an introduction to the field as well as a reference text. It should be noted, however, that the reader will often have to work to unpack some of the theory. This in itself is not a problem, though, because Treiber does provide the relevant references within each chapter to find the resources needed to delve deeper. Although I have some reservations about the structure of this book, overall it is a useful and important work that I will certainly be using in the years to come.

Reviewer:  Harry Strange Review #: CR142161 (1407-0525)
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
General (I.4.0 )
 
Would you recommend this review?
yes
no
Other reviews under "General": Date
Matrix structured image processing
Dougherty E., Giardina C., Prentice-Hall, Inc., Upper Saddle River, NJ, 1987. Type: Book (9789780135656235)
Jul 1 1988
Digital image processing (2nd ed.)
Gonzales R., Wintz P., Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1987. Type: Book (9789780201110265)
Jul 1 1988
Art and design: AI and its consequences
Howard G., John Wiley & Sons, Inc., New York, NY, 1986. Type: Book (9780471909309)
Dec 1 1987
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy