Top 3: Data Science Research For Fashion
1 — Robots Reading Vogue, Yale University
Over a century of Vogue Magazine Covers analyzed at Yale University. Check out the use of color over time, topics talked about over time, algorithmically generate memos in the style of Vogue editor-in-chief, Diana Vreeland.
2 — Cognitive Fashion, IBM Watson
This project was popular in the fashion community when Tommy Hilfiger and IBM partnered with FIT to bring trend insights to student work. The project consists of color analysis, visual search, garment classifiers, complete the look, and partnerships with brands.
3 — Fashion-MNIST and other Projects, Zalando Research
Zalando is a German e-commerce company operating in 17 countries (not currently in the US). They created a popular new standard for training image processing systems called Fashion-MNIST. The original MNIST dataset was a dataset of handwritten digits, while Fashion-MNIST is a dataset of garments. Their research covers more than just this dataset, including: generative models, fashion item encoding, and more!