10
3D Face Recognition
10.1 Introduction
Te automatic recognition of human faces has many potential applications
in various f i elds including security and human–computer interaction. An
accurate and robust face recognition system needs to discriminate between
the faces of dif f erent people under variable conditions. Te main challenge is
that faces, from a general perspective, look similar and their dif f erences can
be very subtle. Tey all have the same structure and are composed of similar
components (e.g. nose, eyes, and mouth). On the other hand, the appearance
of the same face can considerably change under variable extrinsic factors,
e.g. the camera position and the intensity and direction of light, and intrinsic
factors such as the head position and orientation, facial expressions, age, skin
color, and gender. On that basis, face recognition can be considered to be
more challenging than the general object recognition problem discussed in
Chapter 11.
Pioneer researchers initially focused on 2D face recognition, i.e. how to
recognize faces from data captured using monocular cameras. Tey reported
promising recognition results, particularly in controlled environments. With
the recent popularity ofcost-ef f ective 3D acquisition systems, face recognition
systems are startingto benef i tfromthe availability, advantages, andwidespread
use of 3D data. In this chapter, we review some of the recent advances in
3D face recognition. We will f i rst present, in Section 10.2, the various 3D
facial datasets and benchmarks that are currently available to researchers and
then discuss the challenges and evaluation criteria. Section 10.3 will review
the key 3D face recognition methods. Section 10.4 provides a summary and
discussions around this chapter.
3D Shape Analysis: Fundamentals, Teory, and Applications, First Edition.
Hamid Laga, Yulan Guo, Hedi Tabia, Robert B. Fisher, and Mohammed Bennamoun.
(c) 2019 John Wiley & Sons, Inc. Published 2019 by John Wiley & Sons, Inc.