Yuchun's Deep learning study on face color published in JIST

Yuchun’s Deep learning study on face color published in JIST

Yuchun Yan, a Ph.D. candidate of Color Lab has published an article entitled, “Exploring the Facial Color Representative Regions Using the Humanae Images” in the December issue of Journal of Imaging Science and Technology(JIST). The article was submitted to the journal-first track of the Electronic Imaging 2021(EI 2021) conference and accepted for publication. In the study, facial images from the Humanae project were investigated, and the facial color representative regions were explored. During the analysis, the Open CV-libraries were used to detect facial landmarks and to identify ethnic characteristics. At EI 2021, Yuchun will deliver an oral presentation about the study online.

Abstract

It is difficult to describe facial skin color through a solid color as it varies from region to region. In this article, the authors utilized image analysis to identify the facial color representative region. A total of 1052 female images from Humanae project were selected as a solid color was generated for each image as their representative skin colors by the photographer. Using the open CV-based libraries, such as EOS of Surrey Face Models and DeepFace, 3448 facial landmarks together with gender and race information were detected. For an illustrative and intuitive analysis, they then re-defined 27 visually important sub-regions to cluster the landmarks. The 27 sub-region colors for each image were finally derived and recorded in L ∗ , a ∗ , and b ∗ . By estimating the color difference among representative color and 27 sub-regions, we discovered that sub-regions of below lips (low Labial) and central cheeks (upper Buccal) were the most representative regions across four major ethnicity groups. In future study, the methodology is expected to be applied for more image sources.

Link to the article