Decomposing multispectral face images into diffuse and specular shading and biophysical parameters
This work addresses the challenge of detailed skin analysis for applications like computer vision and graphics, though it appears incremental as it builds on existing reflectance modeling.
The authors tackled the problem of estimating diffuse and specular shading and biophysical parameters from multispectral face images by proposing a novel biophysical and dichromatic reflectance model, achieving quantitatively accurate reconstructions and qualitatively convincing decompositions from a single image without complex lighting setups.
We propose a novel biophysical and dichromatic reflectance model that efficiently characterises spectral skin reflectance. We show how to fit the model to multispectral face images enabling high quality estimation of diffuse and specular shading as well as biophysical parameter maps (melanin and haemoglobin). Our method works from a single image without requiring complex controlled lighting setups yet provides quantitatively accurate reconstructions and qualitatively convincing decomposition and editing.