For thousands of years, humans were the only judges of beauty. Even though the criteria for beauty change, for objects and humans both, the judges were always the same, until now. Earlier this year, the first ever beauty contest judged by an AI took place. There were three judges. The Symmetry Master was capable of judging symmetry of facial structure. MADIS was a model which scored people by their similarity to pro-models within their racial group and for sharing common features with famous actors. Lastly, there was RYNKL, a program developed to judge ones wrinkles and level of ageing.
RYNKL has led a successful campaign on Kickstarter, where the pledged $12,187 amounted to almost a double of the intended goal of the project. The RYNKL app is now available at Google Play and App Store. When a selfie is taken through the app, it compares user’s wrinkles in specific areas to a large database, in which the wrinkles have been scored by other people. An AI algorithm then estimates the level of ageing and wrinkleness of user’s face. It also suggests specific treatment and can track changes of the score over time. The score goes from 0-100, zero being the perfect score.
The app is a product of Youth Laboratories, which is a team working under Insilico Medicine – a major big data analysis company. Insilico Medicine is at the same time working of different projects in aligning deep learning with aging and medicine. A paper titled Deep biomarkers of human aging was written by scientists from this company and it got recently published in Aging journal. It outlines features 21 deep neural networks (DNNs), which estimated one’s age and gender based on values recorded from a blood test. The combined level of accuracy for age estimation was 83.5%. The project even has its own website, where you can input your blood test data to get the age estimate.
The implications of this research go far beyond being objective while judging beauty. It is yet another successful use of AI technology, which we can see applied today already. By using two paths to get to the same objective, broadening the database by a crowdfunded campaigns, and at the same time coming up with different methods, Insilico Medicine moves forward fast in this very specific and narrow field.
“One of my goals in life is to minimize unnecessary animal testing in areas, where computer simulations can be even more relevant to humans. Serendipitously, some of our approaches find surprising new applications in the beauty industry, which has moved away from human testing and is moving towards personalizing cosmetics and beauty products”, said Alex Zhavoronkov, CEO of Insilico Medicine, Inc.