Artificial intelligence (AI) continues to make strides in technological advances designed to improve our lives. Speaking of design, one of the latest uses involves a system that can design new clothes based on people’s personal taste using AI. This interesting information came to us from BT in their article, “Let AI dress you as scientists develop a system that designs your wardrobe.”

In an effort to see if machine learning could help the fashion industry, researchers at the University of California, San Diego used AI to build a recommender system to assist online shoppers in clothing purchases.  The team trained an algorithm known as the Siamese convolutional neural network (Siamese-CNN) to make relationships between items selected by the user and similar items in stock.  This laid the foundation of a basic recommender system.  The UCSD team then expanded the capability of the recommender system using a second algorithm, a generative adversarial network (GAN), which can generate images based on user preferences.  The end result is an AI system that not only gives you a “more like this” set of in-stock options, but can actually design new clothing items based on user preferences and system recommendations.

Melody K. Smith

Sponsored by Access Innovations, the world leader in taxonomies, metadata, and semantic enrichment to make your content findable.