Quantitative
characterization of skin appearance is an important but difficult task. The
skin surface is a detailed landscape, with features that depend on many
variables such as body location (knuckle vs. torso), subject parameters
(age/gender/health) and imaging parameters (lighting and camera).
Computational modeling of skin texture has potential uses in many fields and
applications including realistic rendering for computer graphics, robust face
models for computer vision, computer assisted diagnosis for dermatology,
topical drug efficacy testing for the pharmaceutical industry and quantitative
product comparison for cosmetics. In this work, image-based representations of
skin appearance are used in order to have descriptive capabilities without the
need for prohibitively complex physics-based skin models. We present a method
for representing and recognizing different areas of the skin surface that have
visibly different texture properties. Our model takes into account the varied
appearance of the skin with changes in illumination and viewing direction.
In Proceedings of Texture 2002 - The 2nd international workshop on texture analysis and synthesis, pp. 35-41, June 1st, 2002, Copenhagen, Denmark (co-located with European Conference on Computer Vision 2002).
This material is based upon work supported by the National Science Foundation under Grant No. 0092491 and Grant No. 0085864.