Researchers at the University of Texas at Austin and Facebook AI Research have developed a machine learning algorithm to recognize new fashions and then use them to seek out those fashions in a global database of geotagged images.
The team trained the algorithm to recognize styles of clothing in images; it learned to recognize the color, pattern, and garment type of an outfit.
The researchers then allowed the algorithm analyze a dataset containing 7.7 million images of people from 44 cities around the world.
The algorithm was able to identify when and where each style first emerged, and how it eventually spread around the world.
Said the researchers, "Our findings hint at how computer vision can help democratize our understanding of fashion influence, sometimes challenging common perceptions about what parts of the world are driving fashion."
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA