Trends in the implementation of AI in beauty, cosmetology and cosmetology dermatology
Beauty recommendations, which previously a woman could only get by visiting a cosmetologist, are now available by turning to AI. Where is AI used in the beauty industry and does this fact mean that artificial intelligence has learned to understand beauty?
Digital solutions in the beauty industry began to be introduced about 10 years ago. Augmented and virtual reality were the first to be used by retail chains and cosmetics companies in “virtual fitting rooms” and “cosmetology rooms with augmented reality,” allowing beauties to remotely test what a particular hair dye, nail polish, etc. hairstyle, etc.
However, the fundamental difference between AI, taught to taste in beauty and understand the criteria of female beauty, is that it simulates the ability to distinguish between intonations, hints and halftones. Self-educating, he becomes an excellent adviser for women, a useful and tactful “clever friend.”
Prospects for commercialization
In the last two or three years, AI technologies have been successfully commercialized, which, for example, any woman can use through an application on a smartphone. However, without unnecessary hypocrisy, we note that the man too. Here, in my opinion, are the main directions. What are the main directions in which AI applications in the beauty industry are developing today:
Analysis of the skin condition using a photo of the skin area,
medical cosmetology: identification of dangerous/non-hazardous skin tumors from an oncological point of view, identification of allergic processes on the skin,
digital beauty and image advisors and virtual virtual makeup programs,
recommendations on hair, its health and hairstyle,
recommendations for cosmetic products,
creation of new substances and molecules in the cosmetic industry,
interactive training programs and applications.
Let's look at the operating principle of AI using the example of programs in the field of cosmetology dermatology. It is primarily about skin condition analysis and AI recommendations for skin-related procedures.
Applications that analyze the skin can identify dangerous new growths (mole spots, etc.), thereby fulfilling a very important role in the initial diagnosis. And, as you know, prompt diagnosis is very important in this matter. A person may simply be embarrassed to show his spots, moles and pimples, he may not have time to visit a specialist, and this, undoubtedly, lurks the danger of the disease developing to severe forms. Under the hood of all such solutions is image (photo) recognition technology. Training based on image recognition is based on big data associated with data storage of photographs of various areas of the skin, a diagnosis or recommendation for which has already been made. The clustering algorithm produces a result with preliminary recommendations and advice to immediately contact a medical specialist if we are talking about a possible cancerous or precancerous case. Let's look at the functionality of AI in beauty in a little more detail, using examples.
Sample Applications
Neutrogena Skin 360. https://skin360.neutrogena.com/
Main functionality: The “Neutrogena Skin 360” program uses AI to analyze the condition of the skin according to several parameters. The user can scan his face using a smartphone camera or computer webcam. The app uploads it to the server, then uses artificial intelligence to provide personalized care recommendations.
The program examines the skin according to the following parameters:
The presence of graphic “dark circles” (they can be formed by puffiness due to an incorrect diet or improper functioning of internal organs, swelling due to illness or injury, circles under the eyes from an incorrect daily routine, etc.),
Smooth skin
Fine lines on the face, their presence or shape (this includes folds, creases),
Wrinkles,
Skin cleanliness
Skin spots.
Recommendations are the result of extensive AI photography research
AI can advise, having found one or another of the above “defects of ugly lines”: use vitamins, drugs to increase elasticity or smooth out creases and folds, and so on. You can also set a goal you want to achieve (for example, showing the skin tone you like in a photo), and the AI will advise what steps you need to take to achieve the desired cosmetic result.
An example of how Neutrogena Skin 360 works. I tried it on myself as an experiment. But I received some practical recommendations that will be useful in any case:
Strengths: The application is easy to use and does not require any special knowledge. Examines the face quite fully, from different angles (front and on both sides from the side).
Flaws: If your computer/smartphone is not very powerful or the Internet is not very fast, the application may take a relatively long time to produce results.
Now let’s take a brief look at the different applications:
Virtual Makeover
Many brands offer AI-powered apps that allow users to try on virtual makeup before purchasing products. One of the leaders in this matter, which it makes sense to pay attention to first of all, is a French company L'Oreal. Solutions from the startup have become widespread YouCam Mobile.
Recommendations for cosmetic products
Personalized recommendations based on individual preferences and skin type help consumers choose the right beauty products. One of the leaders whose solutions it makes sense to familiarize yourself with first is a Swedish company Foreo.
Medical applications
AI Dermatologist. The application is able to distinguish a tumor that can be removed by cosmetic means from a potential skin cancer.
One of the technologies was developed by scientists from the German Cancer Research Center in Heidelberg. The solutions are based on deep neural networks.
Принятие решений в глубоких нейронных сетях обычно непрозрачно и поэтому непригодно для клинического применения. Здесь мы разрабатываем прозрачную глубокую нейронную сеть для обнаружения меланомы, которая может объяснить ее прогнозы и оценить ее влияние на дерматологов в трехфазном читательском исследовании, - комментирует разработку Katja Hauser, chief data scientist центре в Гейдельберге.
Moral and ethical aspects
People tend to trust smart machines because they consider their point of view, firstly, unbiased. And secondly, based on big data, that is, on a huge array of information, opinions, responses and various images. Their AI is able to group, calculate, find trends and advise, choosing the optimal solution for each.
A modern beauty artist needs to know these trends, understand the strengths and weaknesses of AI technologies and keep abreast of progress in this direction, because the number of clients who will address with the words “but artificial intelligence advised me…” will only grow.
Legal aspects
From a legal point of view, quite a lot of work has already been done to create a legal “coordinate system” between man and artificial intelligence. Being aware of the fundamental laws and definitions is useful and will be useful in the future in any case, even if the student does not plan to use AI “here and now.”
And, of course, the issue of making medical decisions that could potentially affect the client’s health is extremely important: what is the situation with this from a legal point of view? At the moment, all health decisions made by AI are only advisory in nature, preceding a visit to a human specialist. The final decision is up to the doctor. And AI for him is nothing more than a decision support system.
What's next?
It is already possible to predict how AI will develop in the beauty industry in the near future. One one of the vectors is the development of special personal gadgets (for example, skin scanners mounted on a smartphone that highlight and enlarge certain areas of the skin for more detailed photography).
And the main question: AI is becoming more and more powerful, what will this lead to?
The potential and limitations of AI are derived from the technology's core, which is based on big data analysis. Therefore, the prospects for AI in the beauty industry depend on how high-quality the initial data is at the AI’s disposal.
