This AI system can give you fashion advice but bias likely to creep in

US-based researchers have developed a new artificial intelligence (AI) system that can give you fashion vice. Yes, you re that right. Researchers from University of Texas have developed a new AI system named Fashion++ that can suggest tips to make certain outfits more fashionable by simply taking a look at ir picture. According to researchers, this vanced AI system uses visual recognition systems to analyse colour, pattern, texture and shape of garments in an im. AI is so vanced that it also considers where edits will have most impact. Based on that, it would offer alternative outfits to user.

"We thought of it like a friend giving you feedback. It's also motivated by a practical idea: that we can work with a given outfit to make small changes so it's just a bit better," said Kristen Grauman, a professor at University of Texas at Austin in US.

RE | First 'AI' university launched in Abu Dhabi, could India be next?

How AI was trained

Researchers have revealed Fashion++ was trained with help of more than 10,000 ims of outfits shared publicly on online websites, for fashion enthusiasts. Finding ims of fashionable outfits was easy, said gruate student Kimberly Hsiao. However, what was more difficult was finding unfashionable ims, which was proved a challenging task. Researchers mixed pictures of fashionable outfits to create less-fashionable examples and trained system accordingly so that AI could suggest what t to wear.

"As fashion styles evolve, AI can continue to learn by giving it new ims, which are abundant on internet," Hsiao said.

RE | AI predicts effectiveness of antidepressants in patients: Study

Both, Grauman and Hsiao will showcase ir system at International Conference on Computer Vision in Seoul, South Korea next week. Researchers ted that like all AI systems, bias can creep in through data sets for Fashion++. y pointed out that vint looks are harder to recognise as stylish because training ims came from internet, which has been in wide use only since 1990s.

Since user-submitted ims were mostly from rth America, styles from or parts of world don't show up as much, researchers said. Ar challenge is that many ims of fashionable clos appear on models, but bodies come in many sizes and shapes, affecting fashion choices, y said.

(With PTI inputs)