Face Processing: Advanced Modeling and Methods
Format: PDF / Kindle (mobi) / ePub
Major strides have been made in face processing in the last ten years due to the fast growing need for security in various locations around the globe. A human eye can discern the details of a specific face with relative ease. It is this level of detail that researchers are striving to create with ever evolving computer technologies that will become our perfect mechanical eyes. The difficulty that confronts researchers stems from turning a 3D object into a 2D image. That subject is covered in depth from several different perspectives in this volume.
This book begins with a comprehensive introductory chapter for those who are new to the field. A compendium of articles follows that is divided into three sections. The first covers basic aspects of face processing from human to computer. The second deals with face modeling from computational and physiological points of view. The third tackles the advanced methods, which include illumination, pose, expression, and more. Editors Zhao and Chellappa have compiled a concise and necessary text for industrial research scientists, students, and professionals working in the area of image and signal processing.
*Contributions from over 35 leading experts in face detection, recognition and image processing
*Over 150 informative images with 16 images in FULL COLOR illustrate and offer insight into the most up-to-date advanced face processing methods and techniques
*Extensive detail makes this a need-to-own book for all involved with image and signal processing
image/shape-based 7 One exception is the multiview eigenfaces of . Section 1.5: ADVANCED TOPICS IN FACE RECOGNITION 37 methods where no training is carried out. [96, 107, 99, 100] are examples of the ﬁrst class and [132, 106, 98, 97, 117] of the second class. Up to now, the second type of approach has been the most popular. The third approach does not seem to have received much attention. Multiview-based approaches. One of the earliest examples of the ﬁrst class of approaches is the work
computer vision/graphics based approach is presented . This method transfers the appearance change (aging) from a pair of young and senior faces to a new young face to obtain the aged face, or vice versa. The basic idea of this approach is based on the assumption of Lambertian reﬂectance (Equation 9) and the only aging factor considered here is the local surface normal (i.e., the wrinkling). The authors further assume that the change in local surface normal between the young face and senior
of the First Annual INNS Meeting, Boston, p. 515, 1988.  E. I. Hines and R. A. Hutchinson, “Application of multi-layer perceptrons to facial feature location,” IEE Third International Conference on Image Processing and Its Applications, pp. 39–43, July 1989.  G. W. Cottrell and J.Metcalfe, “EMPATH: Face, gender and emotion recognition using holons,” in Advances in Neural Information Processing Systems 3, San Mateo, CA, R.P. Lippman, J. Moody, and D.S. Touretzky (eds.), Morgan Kaufmann,
images relative to the quality of the match between corresponding Gabor jets . The speciﬁc algorithm variants shown are: EBGM USC FERET March 1997. The original CMC curve derived from the similarity matrix generated by University of Southern California (USC) during the FERET evaluations. Inclusion of this curve shows one of the great strengths of evaluations that use and retain similarity matrices: we reconstructed this curve 94 Chapter 3: STATISTICAL EVALUATION OF FACE-RECOGNITION
coordinates of the points in the embedding space. The resulting set of points x1 , . . . , xN in the Euclidean space is called the canonical form of the facial surface [25, 26]. The canonical forms are deﬁned up to a rotation, translation, and reﬂection, and can be therefore treated by conventional algorithms used for rigid surface matching. Figure 5.6 shows an example of a deformable surface (human hand) undergoing isometric transformations, and the corresponding canonical forms of the hand.