Interviews
Facebook's Fashion++ AI system can help dress better
26 Sep '19
2 min read
Pic: Facebook
Outfits can be made more fashionable through style change suggestions using a new Facebook artificial intelligence (AI) system called Fashion++. Designed to offer help on everyday style dilemmas, it may recommend tucking in a shirt, ditching an accessory or swapping a skirt for jeans—easy adjustments that would not force a person to spend on a new purchase.
AI scientists are increasingly finding that technology can amplify human creativity in unexpected ways, said Facebook on its website. Fashion++ is purely a research project, with no immediate plans for related products or features, but it is pointing the way forward, the company clarified.
“While we think such potential applications are exciting, we were also drawn to take on this project because it presents interesting technical problems for AI — such as how to quantify visual compatibility or how to optimize for one property (style) without altering others,” says Kristen Grauman, the research scientist who helped create the tool.
The system’s recommendations will vary, depending on whether it has learned from images of 1920s flapper dresses or 1970s punk rock street styles. The system analyzes those images and applies its learning to decide whether a jeans-and-polo-shirt ensemble is perfect as is or would benefit from a statement necklace.
Computer systems have become quite good at recognising whether a picture shows a dog or a cat, but it is much more challenging to teach them to spot when a sweater clashes with a pair of pants.
The Facebook system also creates its own examples of ‘unfashionable’ outfits. Fashion++ takes the on-trend images it has been given as samples and alters them by swapping out garments, changing patterns, and adjusting the fit. It then uses these artificially generated looks to help teach itself what qualifies as non-fasionable. (DS)
AI scientists are increasingly finding that technology can amplify human creativity in unexpected ways, said Facebook on its website. Fashion++ is purely a research project, with no immediate plans for related products or features, but it is pointing the way forward, the company clarified.
“While we think such potential applications are exciting, we were also drawn to take on this project because it presents interesting technical problems for AI — such as how to quantify visual compatibility or how to optimize for one property (style) without altering others,” says Kristen Grauman, the research scientist who helped create the tool.
The system’s recommendations will vary, depending on whether it has learned from images of 1920s flapper dresses or 1970s punk rock street styles. The system analyzes those images and applies its learning to decide whether a jeans-and-polo-shirt ensemble is perfect as is or would benefit from a statement necklace.
Computer systems have become quite good at recognising whether a picture shows a dog or a cat, but it is much more challenging to teach them to spot when a sweater clashes with a pair of pants.
The Facebook system also creates its own examples of ‘unfashionable’ outfits. Fashion++ takes the on-trend images it has been given as samples and alters them by swapping out garments, changing patterns, and adjusting the fit. It then uses these artificially generated looks to help teach itself what qualifies as non-fasionable. (DS)
Fibre2Fashion News Desk – India
Popular News
Leave your Comments
Editor’s Pick
Therese Premler-Andersson
Textile Machinery Association of Sweden (TMAS)
Folker Stachetzki
Brother Internationale Industriemaschinen GmbH
Andreas Rass
ZIMMER AUSTRIA | Digital Printing Systems
Esteemed Clients
-Ltd..jpg?tr=w-120,h-60,c-at_max,cm-pad_resize,bg-ffffff "Thermore (Far East) Ltd.")
.jpg?tr=w-120,h-60,c-at_max,cm-pad_resize,bg-ffffff "Telangana State Industrial Infrastructure Corporation Limited (TSllC Ltd)")
.jpg?tr=w-120,h-60,c-at_max,cm-pad_resize,bg-ffffff "Taiwan Textile Federation (TTF)")
