Back to articles
Volume: 32 | Article ID: art00005
Beyond Color Correction : Skin Color Estimation In The Wild Through Deep Learning
  DOI :  10.2352/ISSN.2470-1173.2020.5.MAAP-082  Published OnlineJanuary 2020

Estimating skin color from an uncontrolled facial image is a challenging task. Many factors such as illumination, camera and shading variations directly affect the appearance of skin color in the image. Furthermore, using a color calibration target in order to correct the image pixels leads to a complex user experience. We propose a skin color estimation method from images in the wild, taken with unknown camera, under an unknown lighting, and without a calibration target. While prior methods relied on explicit intermediate steps of color correction of image pixels and skin region segmentation, we propose an end-to-end color regression model named LabNet, in which color correction and skin region segmentation are implicitly learnt by the model. Our method is based on a convolutional neural network trained on a dataset of smartphone images, labeled with L*a*b* measures of skin colors. We compare our method with standard skin color estimation approaches and found that our method over-perform these models while removing the need of color calibration target.

Subject Areas :
Views 129
Downloads 57
 articleview.views 129
 articleview.downloads 57
  Cite this article 

Robin KIPS, Loïc TRAN, Emmanuel MALHERBE, Matthieu PERROT, "Beyond Color Correction : Skin Color Estimation In The Wild Through Deep Learningin Proc. IS&T Int’l. Symp. on Electronic Imaging: Material Appearance,  2020,  pp 82-1 - 82-8,

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 2020
Electronic Imaging
Society for Imaging Science and Technology
7003 Kilworth Lane, Springfield, VA 22151 USA