Customer Reviews

Handbook of Face Recognition

5 star:		(1)
4 star:		(1)
3 star:		(1)
2 star:		(3)
1 star:		(0)

 

 

This book contains math and pattern recognition concepts that are not for the faint of heart. However, given that you have some background in pattern recognition and computer vision this book is an invaluable resource for face detection and recognition algorithms and technologies.
Chapter one is a fast paced introduction and goes over face recognition processing from the standpoint of analysis in "face subspaces" along with the technical challenges and solutions that are current.
Chapters two through twelve are all about the many algorithms needed for face recognition and the considerations behind them. First the book tackles how to detect human faces within images. Next the issues of modeling face shape and appearance and tracking faces are explored. Examples of tracker implementations are given. The next few chapters deal with varying conditions and face recognition such as illumination modeling and facial skin color. Chapter seven is a lengthy one on face recognition in subspaces which includes the famous eigenfaces, tensorfaces, and fisherfaces methods. Chapter eight deals with the real-world problem of face tracking and recognition from video and how one does the simultaneous tracking and recognition which is necessary when dealing with video. Chapter 9 is about face recognition in the presence of changing pose and illumination. The concepts of and solutions for normalizing for illumination differences and of modeling pose and illumination are presented. Chapters 10 and 11 round out the algorithmic portion of the book and talk about the more advanced topics of morphable models of faces and facial expression analysis. The math here is quite advanced especially on the topic of morphable models. Chapter 12 is a very short chapter about face synthesis, and seems to be a review of the previous ten chapters on face recognition algorithms along with some new material.
Chapter 13 goes on to the much less mathematical topic of face databases.The various databases of face recognition, detection, and expression analysis are listed and discussed. Chapters 14, 15, and 16 are very accessible and can be understood by both practitioners and by managers wishing to investigate the larger less technical issues of face recogniton. These three chapters are about performance evaluation, psychological and neural perspectives, and face recognition applications respectively.
I would highly recommend this book to individuals who are interested in the implementation of facial recognition systems. Even though it is an edited volume written by a variety of specialists in the field, it has the well-composed feel of a book written by a single author and moves smoothly from one topic to another. You will probably need to supplement your reading if you are trying to design an entire system from scratch, but many of the algorithms needed are detailed in this book. Also, there is an outstanding bibliography for further investigation of various subjects. I have found that a good supplement to the facial expression analysis and the MPEG-4 sections of this book is "MPEG-4 Facial Animation: The Standard, Implementation and Applications" by Pandzic & Forchheimer. It goes into great detail on subjects for which there is simply no practical amount of space in this book.

If you wish to have a useless book with 16 chapters with approximately 50 references per chapter to difficult academic papers (finding these papers is another big problem), with an accompanying text that guide you through these with a lot of photos; than you can buy this one. No mathematical guidence, no backgrounds, no introductory chapters. Just filled with from heaven send formulas, and references. such a complex subject is packed with a 395 pages and named as "handbook". You can't learn anything with this one. One star for photos, and one star for the very superficial text accompanying the huge list of references.

One could probably gleen all the information here by reading a couple of dozen conference proceedings and journal articles, but one of the values of this book is that the information is condensed together into one volume. While some math and algorithms are given, most processes are simply described in general terms so the practitioner may still need to go dig up the referenced articles but at least you'll know where to start digging.

The Internet is constantly changing, so some of the URLs given for image databases in Chapter 13 have already changed. In addition, that chapter fails to mention that not all databases are free and most of them have very restrictive usage licenses that must be agreed to.

Overall I found this text to be a good overview or literature review of recent developments in the technologies of face detection, skin color modeling, face tracking, face modeling, facial expression analysis, and face recognition. As time goes by it will be less and less useful as technology moves forward but for now it is a good starting place for those interested in the field of study.

Don't expect to learn much with this book since the topics are discussed only superficially and with many details laking. In the end, if you really want to understand or implement the methods described in the book, you have to read the original papers that proposed the methods. In this sense, it is a good reference book. Another annoying problem is the difference in notation from one chapter to another.

This book is a compilation of academic papers on biometrics with very little value for the practitioners.

Not quite as good as the handbook of fingerprint technology, but still a fairly good summary of the state of the art in todays facial recognition technology. It seemed like it could have been a little deeper in places such as the area of justice with mugshots.