A Comparative Study of 3-D Face Recognition Under Expression Variations

Abstract

Research in face recognition has continuously been challenged by extrinsic (head pose, lighting conditions) and intrinsic (facial expression, aging) sources of variability. While many survey papers on face recognition exist, in this paper, we focus on a comparative study of 3-D face recognition under expression variations. As a first contribution, 3-D face databases with expressions are listed, and the most important ones are briefly presented and their complexity is quantified using the iterative closest point (ICP) baseline recognition algorithm. This allows to rank the databases according to their inherent difficulty for face-recognition tasks. This analysis reveals that the FRGC v2 database can be considered as the most challenging because of its size, the presence of expressions and outliers, and the time lapse between the recordings. Therefore, we recommend to use this database as a reference database to evaluate (expression-invariant) 3-D face-recognition algorithms. We also determine and quantify the most important factors that influence the performance. It appears that performance decreases 1) with the degree of nonfrontal pose, 2) for certain expression types, 3) with the magnitude of the expressions, 4) with an increasing number of expressions, and 5) for a higher number of gallery subjects. Future 3-D face-recognition algorithms should be evaluated on the basis of all these factors. As the second contribution, a survey of published 3-D face-recognition methods that deal with expression variations is given. These methods are subdivided into three classes depending on the way the expressions are handled. Region-based methods use expression-stable regions only, while other methods model the expressions either using an isometric or a statistical model. Isometric models assume the deformation because of expression variation to be (locally) isometric, meaning that the deformation preserves lengths along the surface. Statistical models learn how the facial soft tissue deforms during expressions based on a training database with expression labels. Algorithmic performances are evaluated by the comparison of recognition rates for identification and verification. No statistical significant differences in class performance are found between any pair of classes.

Surface UP-SR for an improved face recognition using low resolution depth cameras

“…We choose the cosine distance in our experiments as we found it to be the best performing one. This is confirmed by the survey of Smeets et al [20]. .…”

3D Face Analysis: Advances and Perspectives

“…In the other three, ROC curves of All vs. All similarity matrix (i.e. each pair of faces are compared) are usually adopted [98] [102] [33]. Besides, it provides three masks to screen the All vs. All matrix (4007 × 4007) which are of increasing difficulty reflecting the time elapsed between sample acquisition [96] [14] [98] [16].…”

“…Iterative Closet Points (ICP) [40], Iterative Closet Normal Points (ICNP) [140], and Thin-Plate-Spline (TPS) [42], etc. On the other hand, it can [33] be adopted for pose correction, and with landmarks, facial surfaces, in different poses, can be efficiently aligned for further stages, like shape representation.…”

“…Generally speaking, this step seeks a geometric feature space where the differences of face samples are highlighted and distinguished in specific applications. For instance, in 3D FR, shape features are expected to convey person-distinctiveness while remain stable as facial expression varies [33]. In 3D FER, it is desired to generate geometry features that are able to describe subject-independent properties of facial expressions [34].…”

An investigation of matching symmetry in the human pinnae with possible implications for 3D ear recognition and sound localization

“…of people is often required. Therefore, much like fingerprint and face recognition (Smeets et al 2010(Smeets et al , 2012, ear recognition (Chen & Bhanu, 2007;Yan & Bowyer, 2007;Abaza et al 2013) is an active field of research in biometrics (biometric authentication). From this perspective, bilateral symmetry of the human pinnae is also interesting, most obviously because the left ear of an individual can be used as a probe (test subject with unknown identity) in a comparison with a gallery of right ears (database of candidates with known identity) (Abaza & Ross, 2010).…”