laitimes

Observe the business logic behind the "AI + Beauty Industry" digital intelligence application of |: "taking pictures to measure skin quality"

In the last wave of the Internet, beauty was also one of the industries that was greatly affected. "Planting grass" and "beauty UP master" have become key nodes influencing consumer decision-making outside the brand narrative of cosmetics companies. With the "rampant" soft text in "planting grass", the most essential cosmetic ingredients have begun to become an important judgment factor for consumers when making decisions.

Beauty Practice is a third-party beauty ingredient collection platform founded in 2015, and its app is now the #1 beauty app on the iOS Health and Bodybuilding List.

According to the "2021 Beauty Practice Mid-year Consumer Insight Report", skin quality matching has become the first major pain point of scientific skin care. Most people have "knowledge blind spots" in their skin type judgment and skin care methods, which cost them high trial and error. How to determine skin texture and pick out the products that are right for you has been a "key step" in skincare based on consumer cognition.

In the beginning, an average consumer needed to complete a 60-plus question for the American dermatologist Leslie Baumann M.D., who wanted to complete his skin type determination. Sixteen-type skin typing test questionnaire, it takes tens of minutes to complete this questionnaire.

"The acceptance of the 'deep component party' will be relatively high, but as the number of users increases, the simplicity and fun requirements for the process will gradually increase," Yi Ou, CEO of Beauty Cultivation, told the surging news (www.thepaper.cn).

Observe the business logic behind the "AI + Beauty Industry" digital intelligence application of |: "taking pictures to measure skin quality"

In 2020, Beauty Practice cooperated with Megvii to launch the "Take Photos and Measure Skin" function. The running support behind it is Megvii's "face dense key point recognition and color fusion" AI algorithm, which can realize AI intelligent skin quality analysis based on "facial features and skin analysis".

As one of the four domestic AI tigers, Megvii is an artificial intelligence company focusing on the Internet of Things scenario, and has built a software-hard integrated AIoT product system for the three core scenarios of consumer Internet of Things, urban Internet of Things and supply chain Internet of Things, including AIoT operating system, AI redefined hardware and AI redefined industry applications.

When talking about the decision-making path at the beginning of the project, Yi Ou said, "Most of the companies that do the underlying technology are To B and have good technical algorithm advantages, while To C Internet companies are more good at user operation and Internet application level." Our cooperation with Megvii is our own specialty, and the beautiful practice provides megvii's lack of AI training and important links in algorithm upgrading - scene-related images, and Megvii can play a good role in the underlying AI technology. ”

According to the relevant staff, based on Megvii's AI algorithm, it can provide a "virtual makeup trial and intelligent skin measurement" solution to help the beauty and medical beauty industry marketing and retail scene innovation. In terms of intelligent skin measurement, relying on the cooperation between dermatologists and professional institutions in the skin care industry, 23 kinds of skin analysis such as dry oil analysis, sensitivity analysis, pigmentation analysis, aging analysis, acne detection and analysis can be realized, providing users with scientific and accurate personalized suggestions.

Observe the business logic behind the "AI + Beauty Industry" digital intelligence application of |: "taking pictures to measure skin quality"

The cooperation model between the two parties is to take their own strengths, Megvii helps consumers quickly complete the skin condition assessment, and then combines the beauty practice database with 100,000+ brands, 2.8 million+ products, and 28,000+ ingredients to make skin quality matching recommendations, so as to achieve a closed loop of consumption.

Read on