We have hypothesized previously that low contrast and brightness are key optical parameters of attractive, healthy skin and that color is at least as important as topography in affecting these endpoints. Contrast can easily be created by color if a - PowerPoint PPT Presentation

1 / 1
About This Presentation
Title:

We have hypothesized previously that low contrast and brightness are key optical parameters of attractive, healthy skin and that color is at least as important as topography in affecting these endpoints. Contrast can easily be created by color if a

Description:

... 2D colour maps, faces were deformed to match a template grid in ... Image left deformed before 3D rendering. Original image. Final stimulus. Actual age = 12 ... – PowerPoint PPT presentation

Number of Views:77
Avg rating:3.0/5.0

less

Transcript and Presenter's Notes

Title: We have hypothesized previously that low contrast and brightness are key optical parameters of attractive, healthy skin and that color is at least as important as topography in affecting these endpoints. Contrast can easily be created by color if a


1
Visible Skin Color Distribution Plays a Major
Role in the Perception of Age, Attractiveness and
Health in Female Faces
Fink, B., Grammer, K.?, Burquest, M.H. and
Matts, P.J. Department of Anthropology,
University of Göttingen, Germany
?Ludwig-Boltzmann Institute for Urban Ethology,
Vienna, Austria Procter Gamble, Cincinnati,
Ohio, USA Procter Gamble, Egham, Surrey, UK
INTRODUCTION
RESULTS
OBJECTIVE
CONCLUSION
We have hypothesized previously that low contrast
and brightness are key optical parameters of
attractive, healthy skin and that color is at
least as important as topography in affecting
these endpoints. Contrast can easily be created
by color if a homogeneous field is disrupted by
colored features of either sufficient diameter
and / or ratio of adjacent luminance. This is,
indeed, precisely the effect observed in ageing
(particularly photodamaged) human skin, in
features composed of both melanin and hemoglobin.
Skin color distribution, therefore, is
potentially an important visual cue of human
health and beauty and, yet, remains remarkably
unstudied.
To study the single-variable contribution of skin
color distribution to perception of age,
attractiveness and health in female human faces,
independent of facial form, feature and skin
surface topography.
The estimated biological age (aggregated
estimates from all judges for each face) of
facial images ranged from 17.8 to 36.7 years, a
span of some 20 years and there was a highly
significant positive correlation between the
actual biological age of the subjects who
provided facial images and the corresponding
estimated age of their 3D shape-standardised
faces varying only in visible skin colour
distribution (rho .721, plt 0.01, 2-tailed).
Significant negative correlations emerged between
estimated facial age and the global face
attributes (attractive rho -.527, plt .01
healthy rho -.520, plt .01 youthful rho
-.860, plt .01). In summary, therefore, we can
conclude that skin colour distribution alone,
independent of facial form, feature and skin
surface topography can influence perceived age
within a range of 20 years. Furthermore, it also
appears to influence significantly perceived
attractiveness, youth and health.
Strategies to improve the appearance of ageing
skin, therefore, need to focus not only contrast
created by form and topography, but also that
created by color distribution and the chromophore
targets responsible for this.
b
a
In this stage, facial features (e.g. pupils,
mouth gap, etc.) were standardized geometrically
by fitting these to fixed addresses within the 2D
template. To generate 3D facial stimuli from 2D
colour maps, faces were deformed to match a
template grid in order to fit on a
shape-standardised wire-frame mesh. In the final
rendering process, these corrected 2D maps were
fitted to the wire-frame mesh, akin to a virtual
skull. In this process, we added standardised
facial features (eyes, nose, mouth, ears, hair,
etc.) such that the resulting dataset comprised
169 3D head / face stimuli, standard in every
respect apart from the subjects original skin
colour distribution. An example of this process
is shown in the panel opposite. These stimuli
were shown blind to 430 members of the public,
aged 13-76, in Germany and Austria using
calibrated monitors. Participants were requested
to estimate the biological age of each face using
a single-step scale ranging from 10 to 60 years.
In addition, participants were asked to rate each
face for a total of 15 attributes using a
10-point rating scale combining aspects of
perceived attractiveness and health and apparent
skin condition.
(a) 2D template (b) 3D wire-frame mesh (front and
right)
(a) Original frontal image (b) lines / furrows
removed
METHODS
169 Caucasian women aged between 10 and 70 were
imaged from front, left and right views using a
custom high-resolution digital imaging system.
The use of cross-polarised lighting eliminated
fine surface texture in this imaging stage. The
resulting images were processed using a new,
unique series of digital manipulations to create
stimulus heads where skin color distribution
was the only remaining variable Left and right
sides were grafted onto the frontal image and
then a cloning technique was used to remove any
contrast attributable to low-frequency
topographical features (lines / furrows, etc).
2D color maps were than created by fitting the
resulting image to a standard 2D template.
Image above (b) fitted to 2D template
Image left deformed before 3D rendering
Original image
Final stimulus
30
28
26
Estimated Age
24
22
20
References Fink, B. Grammer, K. Thornhill,
R. Human (Homo sapiens) facial attractiveness in
relation to skin texture and color. Journal of
Comparative Psychology, 115(1), 92-99,
2001 Matts, P.J., Understanding and measuring
the optics that drive visual perception of skin
appearance, in The essential stratum corneum,
Marks, Leveque, Voegeli (eds), Martin Dunitz,
London, 213-222, 2002
Actual Age (5 year groups)
Actual age 12 Estimated age 20
Actual age 42 Estimated age 24
Actual age 55 Estimated age 31
LSD ANOVA plot (means and confidence intervals)
of actual chronological age (5 year cohorts) vs
estimated age
Examples of final rendered stimuli with actual
and estimated ages. Note the range of appearance
attributable to colour distribution alone.
This work was funded by PG Beauty
Write a Comment
User Comments (0)
About PowerShow.com