Artificial Intelligence is bringing about a revolution in all possible industries, and fashion is no exception. In future, AI might take the job of fashion designers that actually learns your fashion style and become your personal shoppers.
Researchers at the University of California at San Diego have partnered with Adobe Research to create this new fashion-oriented AI that will not only learns about your personal fashion style but can also make recommendations (generate realistic images) based on your preferences. So rather than simply finding clothes (that already exist) which would suit the person’s tastes, it can actually invent new items of clothing perfect for them personally.
This new AI technology could be used not only on a personal level to create individual clothing, but on a wider scale, it could also be used to help predict future fashion trends. In a paper published on the ArXiv, the researchers have outlined a way for the system that could let retailers create personalized pieces of clothing, or could even be used to help predict broader fashion trends.
So how does this work?
According to researchers, there are two different algorithms powering this AI solution. One is a Convolutional Neural network (CNN) and the other is Generative Adversarial Network (GAN).
First, the researchers trained CNN to learn and classify user preferences for specific items, using purchase data scraped from Amazon in six different categories, including footwear, tops, and pants for both women and men. This type of recommender model is common in the online retail world, usually showing up in an “Other items you might like” area at the bottom of a page. This algorithm is not entirely new in the world of online retail, it is much like current systems which offer you “You may also like” sections when shopping online.
The second algorithm is where it gets really smart, generating clothing suggestions to consumers. In the case of GANs, two neural networks are trained with the same set of data. One of them is assigned with the task to create fake images of the data and the other verifies whether the images are real or not. This technology is also not new, as GANs were invented a few years ago in 2014 by Ian Goodfellow.
The researchers used the information gathered by the CNN algorithm as the data set for GAN which then created multiple images for each user. You can read the research paper published on ArXiv here.
From a commercial perspective, if this new GAN-based AI technology ends up on e-commerce websites and online recommendation systems, it could do wonders for them. It could help retailers discover new items that people want without either party even realizing it. At this stage, the system is in its early stages of development and can only produce simple shapes but the future could be very different. Companies like Amazon and Alibaba are already working on their versions of fashion designer AIs.
Watch out designers, we could soon have the new fashion trends dictated by artificial intelligence.