June 10, 2020
The Fashion IQ Challenge of the CVPR2020, is the second competition hosted by IBM Research.
Fashion IQ is a competition that recommends the most suitable clothes according to the user's natural language (people's daily language) search. When the user enters a modification into a natural language in the proposed product image, it reflects this and recommends a new product image.
Designing an algorithm that reflects the user's repetitive natural language feedback well is a key technology in developing an interactive search/recommendation system.
The Ripple A-Seoul National University team (Professor Gunhee Kim, graduate student Youngjae Yu and Jongseok Kim, Ripple AI researcher Lee Seunghwan Lee) ranked first among the 44 teams.
They were the runner-up at the first competition held at ICCV2019.
The presentation and awards ceremony was held on June 19th through the CVPR workshop.
Further details can be found at https://sites.google.com/view/cvcreative2020/home.
Relating paper can be found at Paper.