Abstract
Measuring facial traits by quantitative means is a prerequisite to investigate epidemiological, clinical, and forensic questions. This measurement process has received intense attention in recent years. We divided this process into the registration of the face, landmarking, morphometric quantification, and dimension reduction. Face registration is the process of standardizing pose and landmarking annotates positions in the face with anatomic description or mathematically defined properties (pseudolandmarks). Morphometric quantification computes pre-specified transformations such as distances. Landmarking: We review face registration methods which are required by some landmarking methods. Although similar, face registration and landmarking are distinct problems. The registration phase can be seen as a pre-processing step and can be combined independently with a landmarking solution. Existing approaches for landmarking differ in their data requirements, modeling approach, and training complexity. In this review, we focus on 3D surface data as captured by commercial surface scanners but also cover methods for 2D facial pictures, when methodology overlaps. We discuss the broad categories of active shape models, template based approaches, recent deep-learning algorithms, and variations thereof such as hybrid algorithms. The type of algorithm chosen depends on the availability of pre-trained models for the data at hand, availability of an appropriate landmark set, accuracy characteristics, and training complexity. Quantification: Landmarking of anatomical landmarks is usually augmented by pseudo-landmarks, i.e., indirectly defined landmarks that densely cover the scan surface. Such a rich data set is not amenable to direct analysis but is reduced in dimensionality for downstream analysis. We review classic dimension reduction techniques used for facial data and face specific measures, such as geometric measurements and manifold learning. Finally, we review symmetry registration and discuss reliability.
Highlights
When Active shape models (ASMs) are implemented for 2D images, a projection step from 3D to 2D can be used to apply these models on surface data
The degree of symmetry is an important property of a face having a connection with attractiveness (Scheib et al, 1999; Leyvand et al, 2008) and an impression of dysmorphia which in turn is linked with genetic syndromes (Winter, 1996; Thornhill and Møller, 1997)
Both empirical (Molinaro et al, 2005; Boehringer et al, 2011a) and theoretical (Bai and Silverstein, 2010) considerations imply that Principal component analysis (PCA) has the largest influence on reproducibility by contributing variability into the analysis that is larger than that induced by measurement error
Summary
The face plays an important role in human interaction and is scientifically researched with respect to many disciplines including genetic control (Boehringer et al, 2011b; Liu et al, 2012; Paternoster et al, 2012; Adhikari et al, 2016; Cole et al, 2016; Shaffer et al, 2016; Lee et al, 2017; Claes et al, 2018), psycho-social impact (Scheib et al, 1999; Leyvand et al, 2008), archaeology (Stenton et al, 2016), forensic reconstruction (Short et al, 2014), relation with medical conditions (Hammond et al, 2005; Boehringer et al, 2006, 2011a; Vollmar et al, 2008; Wilamowska et al, 2012), and facial identification (Wiskott and Von Der Malsburg, 1996; Schroff et al, 2015; Sun et al, 2015). We cover automatic methods that can derive quantitative values, such as distances, from given facial raw data as represented by 2D photographs or 3D surface scans. Such quantities are of interest in genome wide association studies (GWASs), syndrome classification, and other prediction settings. By facial alignment we mean that some or all points of a given face can be transformed to points on another face while retaining their meaning in terms of landmarks, thereby establishing correspondence between landmarks. Aligned landmark data is rarely analyzed directly Rather they are processed further using either pre-specified transformations or dimension reduction techniques. Symmetry is quickly discussed as a face specific application and some open problems are mentioned
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
