Facial image analysis for skin care: how cosmetics maker C’BON uses AI to develop beauty technology

*This is a machine-translated text of the original article.

C’BON Cosmetics Co Ltd announced on March 11, 2020 that they developed AI technology in collaboration with Keio University to accurately capture the position of the eyebrows, eyes, nose, and mouth.

The company has developed a unique skin counseling system at its own facialist salon and analyzes images of the entire face.

How AI technology accurately captures facial features

The aim of this study was to improve the accuracy and efficiency of further image analysis.

The technique explores the area containing each part of the face (eyebrow, eye, nose, and mouth) in the image and sets the reference point for each part. Determine the region of each part based on its reference point coordinate. In addition, feature point coordinates are determined within each part region.

These procedures allowed us to clarify the feature point extraction algorithms for frontal and transverse faces and to perform precision assessments.

In conducting image analysis, the company must accurately capture the location of the facial parts (eyebrows, eyes, nose, and mouth) in the image, and the facialist is now performing this task manually. Therefore, subtle errors in position can occur depending on the hand movement.

As a result of joint research with Keio University’s Department of Science and Engineering to further improve the accuracy of analysis, we succeeded in automating the extraction of feature points (*) using AI technology.

In addition, this technology will be used in the future.
The company is planning to introduce it into beauty log advice navigators offered by facialist salons and others.

(*)Marks placed at specified positions such as the head and tail of the eye in the image

Original Text: https://ampmedia.jp/2020/03/11/cbon/

モバイルバージョンを終了