Amazon Shopping AI Lets Consumers Try On Digital Outfits, Makes Fashion Suggestions

Amazon could be planning an artificial intelligence-driven virtual shopping service, according to documents from a presentation planned for the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), VentureBeat reports.

The service would be multi-layered and work through several different algorithms. Users will be able to use the first algorithm to hone search queries by describing variations on a product image, and another algorithm will suggest items that match with things a customer has already picked out.

The third algorithm will synthesize an image of a model wearing clothes from the user's search, which will show the way the clothes work as an outfit.

At Lab126, Amazon's hardware facility — which has put out products like Fire TV, Kindle Fire and Echo — researchers have also put together a virtual try-on system, called Outfit-VITON. That will attempt to show what an outfit might look like on an image of a person. The system will utilize a generative adversarial network (GAN), which allows for the distinguishing of a generated item from a real image, Amazon says.

Outfit-VITON will work by using a shape-generation model that uses a template for the final image — and however many reference images a user wants to provide. The program then does its best estimation of the user's body type and color, to provide as close a guess as possible as to how an outfit would look on them in real life.

The researchers say online clothes shopping would be a convenient new way to go about it, but did note the inherent flaw of the model, saying the overall effect would be limiting.

“[O]nline shopping does not enable physical try-on, thereby limiting customer understanding of how a garment will actually look on them,” the researchers wrote. “This critical limitation encouraged the development of virtual fitting rooms, where images of a customer wearing selected garments are generated synthetically to help compare and choose the most desired look.”

The coronavirus pandemic, with its disruption of normal in-person shopping, has innovators looking for new ways of continuing to do the things Americans always have in the past, such as new ways of watching movies.

And Amazon's ideas are in-line with others being tried as of late like augmented reality shopping trips.



The How We Shop Report, a PYMNTS collaboration with PayPal, aims to understand how consumers of all ages and incomes are shifting to shopping and paying online in the midst of the COVID-19 pandemic. Our research builds on a series of studies conducted since March, surveying more than 16,000 consumers on how their shopping habits and payments preferences are changing as the crisis continues. This report focuses on our latest survey of 2,163 respondents and examines how their increased appetite for online commerce and digital touchless methods, such as QR codes, contactless cards and digital wallets, is poised to shape the post-pandemic economy.