Testing the Utility of a Data-Driven Approach for Assessing BMI from Face Images

Wolffhechel, Karin Marie Brandt; Hahn, Amanda; Jarmer, Hanne Østergaard; Fisher, Claire I.; Jones, Benedict C.; DeBruine, Lisa

doi:10.1371/journal.pone.0140347

Cited by 27 publications

(30 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using different regression methods based on cross-validated sets of PCs, Wolffhechel et al [15] were able to account for 23% of BMI variation by facial shape, which is in the range of our results. Strikingly, they found that facial texture predicts 27% of variation in BMI, while we could predict only about 3% after cross-validation.…”

Section: Discussionsupporting

confidence: 78%

“…Wolffhechel and colleagues [15] predicted BMI by face shape and texture using more conventional multiple regression approaches. As they require substantially more cases than variables, we performed the regressions using smaller numbers of PCs.…”

Section: Resultsmentioning

confidence: 99%

“…This difference may result from the different regression methods and sample sizes, but more likely, it results from the differences in image registration. Most of the landmarks in our study as well as that in Wolffhechel et al [15] are semilandmarks, that is, they are placed on homologous curves (e.g., the facial outline), but their positions along this curve are arbitrary. We use the sliding landmark algorithm [23,24], which slides the semilandmarks along their curves in order to minimize “bending energy”–a measure of local shape deformation [27]–between the individuals’ landmarks and their sample average.…”

Section: Discussionmentioning

confidence: 96%

“…For example, using a principal component model, Wolffhechel et al [15] could predict 23% of variation in BMI by face shape, 27% by face color, and 30% by combining both face shape and color. Some authors related perceived BMI to judgments of female mate value, such as attractiveness and (perceived) health.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

BMI and WHR Are Reflected in Female Facial Shape and Texture: A Geometric Morphometric Image Analysis

et al. 2017

View full text Add to dashboard Cite

Facial markers of body composition are frequently studied in evolutionary psychology and are important in computational and forensic face recognition. We assessed the association of body mass index (BMI) and waist-to-hip ratio (WHR) with facial shape and texture (color pattern) in a sample of young Middle European women by a combination of geometric morphometrics and image analysis. Faces of women with high BMI had a wider and rounder facial outline relative to the size of the eyes and lips, and relatively lower eyebrows. Furthermore, women with high BMI had a brighter and more reddish skin color than women with lower BMI. The same facial features were associated with WHR, even though BMI and WHR were only moderately correlated. Yet BMI was better predictable than WHR from facial attributes. After leave-one-out cross-validation, we were able to predict 25% of variation in BMI and 10% of variation in WHR by facial shape. Facial texture predicted only about 3–10% of variation in BMI and WHR. This indicates that facial shape primarily reflects total fat proportion, rather than the distribution of fat within the body. The association of reddish facial texture in high-BMI women may be mediated by increased blood pressure and superficial blood flow as well as diet. Our study elucidates how geometric morphometric image analysis serves to quantify the effect of biological factors such as BMI and WHR to facial shape and color, which in turn contributes to social perception.

show abstract

Section: Discussionsupporting

confidence: 78%

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 96%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

BMI and WHR Are Reflected in Female Facial Shape and Texture: A Geometric Morphometric Image Analysis

et al. 2017

View full text Add to dashboard Cite

show abstract

“…These three measures do not necessarily capture all aspects of face shape that relate to weight perceived from faces. It is also possible to use an empirical approach to derive more global measures of facial shape that characterize facial adiposity using principal component analysis (PCA) of landmarks capturing the structure of two-dimensional (2D) facial images [32] or the entire surface of three-dimensional (3D) faces [33]. It remains to be shown how such measures relate to perceived health and measured health.…”

Section: (C) Facial Adipositymentioning

confidence: 99%

Perception of health from facial cues

Henderson

Holzleitner

Talamas

et al. 2016

Phil. Trans. R. Soc. B

View full text Add to dashboard Cite

Impressions of health are integral to social interactions, yet poorly understood. A review of the literature reveals multiple facial characteristics that potentially act as cues to health judgements. The cues vary in their stability across time: structural shape cues including symmetry and sexual dimorphism alter slowly across the lifespan and have been found to have weak links to actual health, but show inconsistent effects on perceived health. Facial adiposity changes over a medium time course and is associated with both perceived and actual health. Skin colour alters over a short time and has strong effects on perceived health, yet links to health outcomes have barely been evaluated. Reviewing suggested an additional influence of demeanour as a perceptual cue to health. We, therefore, investigated the association of health judgements with multiple facial cues measured objectively from two-dimensional and three-dimensional facial images. We found evidence for independent contributions of face shape and skin colour cues to perceived health. Our empirical findings: (i) reinforce the role of skin yellowness; (ii) demonstrate the utility of global face shape measures of adiposity; and (iii) emphasize the role of affect in facial images with nominally neutral expression in impressions of health.

show abstract

Body mass index, facial width‐to‐height ratio, and perceived formidability in female Ultimate Fighting Championship (UFC) fighters

Palmer-Hague

Twele

Fuller

2018

Aggressive Behavior

View full text Add to dashboard Cite

Both facial width-to-height ratio (fWHR) and body mass index (BMI) have been associated with aggressive behavior in women but how they influence perception of their potential threat remain unclear. Here, we assessed the effects of fWHR and BMI on perceived formidability from faces of 42 female Ultimate Fighting Championship (UFC) fighters. In study 1, BMI, but not fWHR, positively predicted participants' ratings of aggressiveness and fighting ability from facial photographs. In study 2, both high fWHR and high BMI composite faces were rated as more aggressive, tougher, and more likely to win a fight than low fWHR and low BMI composite faces, respectively. Further analyses revealed that the high BMI composite face was rated as more aggressive and tougher than the high fWHR composite face. Taken together, these results suggest that compared to fWHR, BMI may be a more salient cue to women's formidability.

show abstract

Testing the Utility of a Data-Driven Approach for Assessing BMI from Face Images

Cited by 27 publications

References 30 publications

BMI and WHR Are Reflected in Female Facial Shape and Texture: A Geometric Morphometric Image Analysis

BMI and WHR Are Reflected in Female Facial Shape and Texture: A Geometric Morphometric Image Analysis

Perception of health from facial cues

Body mass index, facial width‐to‐height ratio, and perceived formidability in female Ultimate Fighting Championship (UFC) fighters

Contact Info

Product

Resources

About