TRY-ON
News
Oct 20, 2020

Perfect Corp. Announced Winners of the 3rd Annual “AI Meets Beauty” Challenge at the Virtual ACM Multimedia Conference

Perfect Corp. Announced Winners of the 3rd Annual “AI Meets Beauty” Challenge at the Virtual ACM Multimedia Conference

This year’s global competition spotlighted advanced AI innovations focusing on beauty and personal care product recognition.

New YorkOctober 20, 2020Perfect Corp., the leading AI + AR beauty tech solutions provider, announced the winning teams of the third annual “AI Meets Beauty Challenge” at the ACM Multimedia online conference on October 16th. The “AI Meets Beauty” challenge is an annual competition aimed at cultivating smart retail solutions that improve customer experience in the beauty industry. A total of 33 teams from 10 different countries participated in this year’s challenge, which focused on beauty and personal care product recognition. The top three prizes were awarded to the teams from University of Science and Technology of China, Guangdong University of Technology, and Nanyang Technological University.

The winning team presented a new technique called AMAC (Attention-Based Maximum Activation Convolution) that helps accurately extract the target beauty product from the given image for further identification. The innovation resulted in over 30% accuracy improvement compared to last year’s results. The strenuous yet constructive competition produced six academic papers that will be published at ACM Multimedia Conference 2020. This outcome further reinforces the scientific and innovational significance of the “AI Meets Beauty” challenge.

Participants were given a real-world image containing one beauty or personal care item and they had to develop an AI algorithm that would accurately match the real-world example of the given item to the same item in the Perfect-500K data set. The task has proven to be quite difficult considering the limitation that only images from e-commerce sites were available in Perfect-500K and no real-world examples were provided in advance.

“Congratulations to the winners of the 2020 “AI Meets Beauty” challenge for developing ground-breaking and disruptive beauty AI solutions. We are beyond impressed with everyone’s submissions,” said the CEO and founder of Perfect Corp., Alice Chang. “The finalists have proven their power to innovate with different but unique AI solutions for the beauty industry. We are happy to see so many young talents who will undoubtedly drive the beauty tech industry forward and transform the consumer beauty journey.”


About Perfect Corp.

With over 850 Million downloads globally, Perfect Corporation is dedicated to transforming how consumers, content creators and beauty brands interact together through AI and AR technologies. Our experienced team of engineers and beauty aficionados are pushing the frontiers of technology to create the beauty platform of the future – a fluid environment where individuals express themselves, learn the latest about fashion and beauty, and enjoy instant access to the products from their favorite brands. Further information about Perfect Corp. can be found at perfectcorp.com and LinkedIn.

Press Contacts

Perfect Corp. official website: https://www.perfectcorp.com/

Perfect Corp. on LinkedIn: https://www.linkedin.com/company/perfect-corp/

Corporate: Tony Tsai at press@perfectcorp.com or by phone at: +886-2-8667-1265, ext. 2167

USA: Lindsay London at contact_pr@perfectcorp.com or by phone at +1 (212) 929-7979

Japan: Ryusho Hosaka at contact_pr_jpn@perfectcorp.com or by phone at: +81-3-5875-6651

China: Celine Lo at celine_lo@perfectcorp.com or by phone: + 86 1582 112 8182 

Europe: Jessica Thiant at contact_pr_fr@perfectcorp.com

 

All companies and product names mentioned herein are for identification purposes only and are the sole property of their respective owners. Copyright © 2020 Perfect Corp. All rights reserved.



By using the website, you agree to our use of cookies. Head to our cookie statement to learn more about cookies and manage cookies on this website.