Book Reviews

Face Processing: Advanced Modeling and Methods

J. Electron. Imaging. 15(4), 049901 (November 27, 2006). doi:10.1117/1.2397688
History: Published November 27, 2006
Text Size: A A A

Open Access Open Access

The editors of this book, Wenyi Zhao and Rama Chellappa, deliver a compendium of the state of the art in face processing. They assemble chapters written by leading researchers in various aspects of the field into a book that contains 21 chapters and is divided into three major sections: the basics, face modeling, and advanced methods.

As the section heading suggests, “Part I: The Basics” introduces many of the fundamental topics and issues in face processing. It contains three chapters entitled “A Guided Tour of Face Processing,” written by the editors; “Eigenfaces and Beyond,” authored by M. Turk; and “Introduction to the Statistical Evaluation of Face-Recognition Algorithms,” by J. R. Beveridge et al. The first chapter introduces and distinguishes the tasks of face detection, feature extraction, and face recognition. It discusses broad approaches to each of these problems and introduces the complications created by variations in pose, illumination, and expression. Principle component analysis (PCA) is a tool applied to many aspects of face processing. Turk describes its use as the basis of eigenfaces, and thus lays the foundation for its use throughout the book. The third chapter is somewhat less general than the preceding chapters. The authors compare the performance of face recognition systems based on PCA with those based on linear discriminant analysis (LDA). In doing so, they introduce standard face datasets and measures of performance. A nice addition to this section would have been Chapter 8: Face Recognition by Humans (currently found in Part II).

The second section, “Part II: Face Modeling,” is further divided into two subsections. The first contains four chapters addressing computational aspects of face modeling and recognition. The second contains three chapters on psychophysical aspects of human perception and cognition. The computational subsection includes chapters introducing morphable face models, expression invariant representation and recognition, building 3-D models from video, and face modeling by information maximization. The psychophysical subsection includes several chapters investigating face recognition by humans, predicting human face recognition performance, and the distribution of face and object recognition within the brain. The discussions within this subsection note correlations and contrasts between the recognition performance of humans and machines.

The third section, “Part III: Advanced Methods,” delves deeper into more specific subject areas. Several chapters address complications resulting from variations in illumination. Other chapters address working with individual facial components as opposed to holistic face representations, with applications toward modeling, recognition, tracking, and occlusion detection. This section also includes chapters on real-time facial processing, thermal imaging, and multimodal biometrics.

Since the book is a compilation of chapters written by many different authors, the writing style inevitably varies from chapter to chapter. However, while some chapters are presented more formally than others are, all are well written. Occasionally there is some redundancy between chapters, but this may allow readers to target chapters of interest without needing to refer to previous sections. Chapters typically provide a firm mathematical basis for the technologies they describe, and end with a wealth of references for readers seeking more details of particular implementations. The chapters are formatted with a clear, consistent style. Figures within the body of the text are reproduced as good quality monochromatic half-toned images. The center of the book contains eight pages of full-color plates.

Research into face processing is broad and varied. Thus, there are many topics not covered by this book or mentioned only in passing. Yet, this book delivers a good overview of important topics addressed by current research. The early chapters provide an excellent introduction for those new to face processing or those peripherally involved who need a good general understanding. The latter chapters provide a good reference for those working in the field with ample references to key papers. Much of the contents of this book overlaps that of the book Handbook of Face Recognition, edited by Stan Z. Li and Anil K. Jain [ISBN: 038740595X, Springer (2005)]. However, the complementary content and the excellent presentation makes this book highly recommended, and a valuable addition to your bookshelf (at least until one of your colleagues spots it).

Peter Stubler is a principle scientist with the Eastman Kodak Company in the field of visual information management. His current research interests are semantic content-based image retrieval and face recognition for personal image databases.

Peter O.StublerIndividualAuthor

Citation


"Face Processing: Advanced Modeling and Methods", J. Electron. Imaging. 15(4), 049901 (November 27, 2006). ; http://dx.doi.org/10.1117/1.2397688


Figures

Tables

References

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Related Book Chapters

Topic Collections

Advertisement
  • Don't have an account?
  • Subscribe to the SPIE Digital Library
  • Create a FREE account to sign up for Digital Library content alerts and gain access to institutional subscriptions remotely.
Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).
Access This Proceeding
Sign in or Create a personal account to Buy this article ($15 for members, $18 for non-members).
Access This Chapter

Access to SPIE eBooks is limited to subscribing institutions and is not available as part of a personal subscription. Print or electronic versions of individual SPIE books may be purchased via SPIE.org.