Studies in Skin Colors

Download Report

Transcript Studies in Skin Colors

Simon Fraser University
Computational Vision Lab
Lilong Shi, Brian Funt and Tim Lee

Studies of factors affecting skin colour

Simple and linear model of skin

Modelling Skin appearance under lights

Applications:
 Estimate melanin and hemoglobin concentrations
 Correct imaged skin tones for lighting conditions
Skin tone
correction
Tone
correction
Preserve
melanin
Melanin/Hemoglobin separation

Appearance of human skin determined by
 Biological factors
▪ pigmentation, blood microcirculation, roughness, etc..
 Viewing conditions
▪ Inducing lights
 Acquisition devices
▪ Cones in retina, RGB sensors of CCD digital cameras

Two-layered Skin Model [2]
 Epidermis Layer: melanin absorbance
 Dermis Layer: hemoglobin absorbance

A layer has properties of an optical filter

Various skin colour <= melanin + hemoglobin
 Genetic: Race
 Temporary:
▪ Exposure to UV
▪ Hot bath


Mixture varying by 2 independent factors
Analyse melanin and hemoglobin factors

Estimate melanin and hemoglobin concentration

Independent Component Analysis (ICA)
– Statistical technique for revealing “hidden” factors
– To “unmix” or “separate” signals composed of
multiple sources
– Independent and linear mixing
– Related to Eigen-vector analysis
Original Source Signals
Mixing
Observed Signals
v1
70%
s1
0%
30%
20%
s2
v2
80%
100%
v3
s
×
A
=
v
Melanin
Melanin
Hemoglobin
Hemoglobin
Skin
samples

Typical skin spectrum
 Visible wavelength 400nm – 700nm

Extract skin bases from observed spectrum by ICA
ICA
(left) 33 skin spectrum
after normalization;
(right) two independent
basis spectrum – the
melanin and
hemoglobin, and the
spectrum of
chromophores other
than melanin and
hemoglobin pigments.

Arbitrary skin spectrum can be approximated
mσm  hσh  c
m , h
are variables
constru
 Human vision
▪ 3 types of Photoreceptors
 L, M and S Cones
 Digital Cameras
▪ 3 sensors
 Red, Green, and Blue
 Reflectance spectrum recorded by 3 sensors
=> three values (R, G, B) for a skin colour
pixel mσ m  hσ h  c
Given a pixel from
skin, compute m , h
by projecting
log(R,G,B) onto σ m , σ h
Possible skin colours
lie within plane
mσm  hσh  c
m
Melanin
Image
h
Hemoglobin
Image
Input Image [3]
- Inverse melanin concentration
- Inverse hemoglobin concentration

Skin appearance greatly affected by lights

Reveal true skin colour by removing illum.

Common lights blackbody radiation
 e.g. tungsten/halogen lamps, sunrise/sunset, etc
 Varying colour temperature T
▪ Redish -> white -> bluish


Colour: illumination times reflectance
In log space, multiplication => addition:
Π( m , h , )  mσ m  hσ h  τω  c
Illum. basis

In practice
Π( m , )  mσm  τω

Drop hemoglobin basis
▪ Small angle between Illum and hemoglobin axes

Ignore brightness

Skin colour varying by T and
m


384 real skin reflectances times
67 real light sources

=> 25728 samples
Π( m , )  mσm  τω

Skin tone correction example (UOPB DB [4])
16 different
illumination +
camera settings
Tone
correction
Preserve
melanin
20
• Skin tone correction example (UOPB DB [4])

Skin colour modelling:
 Melanin and Hemoglobin concentration
 Linear model in logarithm space
 Estimation by Independent Component Analysis

Skin appearance + Light modelling:
 Estimates light source
 Preserves skin colour by melanin value

Applied to digital images from CCD cameras

[1] Shi, L., and Funt, B., "Skin Colour Imaging That Is Insensitive to
Lighting," Proc. AIC (Association Internationale de la Couleur) Conference
on Colour Effects & Affects, Stockholm, June 2008

[2] Angelopoulou, E., Molana, R., and Daniilidis, K. “Multispectral skin
color modeling,” In IEEE Conf. on Computer Vision and Pattern
Recognition, volume 2, pages 635-642, Kauai, Hawaii, Dec. 2001.

[3] Shimizu, H., Uetsuki, K., Tsumura, N., and Miyake, Y. Analyzing the
effect of cosmetic essence by independent component analysis for skin
color images. In 3rd Int. Conf. on Multispectral Color Science, pages 65-68,
Joensuu, Finland, June 2001.

[4] Martinkauppi, B. “Face color under varying illumination-analysis and
applications,” Ph.D. Dissertation, University of Oulu, 2002.