Over the course of 10 weeks, students at the Fashion Institute
of Technology in New York City explored IBM’s Cognitive
Fashion tools and developed innovative new designs and marketing strategies for Tommy Hilfiger; a major partner to the project who contributed approximately 15,000 catalog images from the past three years. IBM Researchers initially used Tommy Hilfiger product image data to generate a report on popular and trending color palettes. The report helped spark the students creative process but they leveraged IBM’s Cognitive Fashion tools to visually explore styles and trending silhouettes, and even use a tool which generated novel prints from scratch.
“The machine-learning analysis gave us insights about the
Tommy Hilfiger ‘DNA’ that we couldn’t begin to consume or understand with the human mind.” - Michael Ferraro, executive director of the FIT/Infor Design and Tech Lab. . We used publicly available fashion images along with visual recognition techniques to analyze and predict the color and print pattern trends in fashion which were used by the designer to choose the color palette for his designs. We are building a cognitive agent that can analyze current fashion trends from multiple sources (catalogs, articles, blogs, images, social media) and forecast future fashion trends. For consumers who are unsure of exactly what they want, or they’ve seen something that’s inspired them in the real world, but can’t put it into words, visual search will help accelerates the discovery process. A user clicks on an existing apparel or accessory image in the product catalog and Visual Browse will fetch and display other similar looking products from the catalog. Visual Browse does not need any meta information as searches are performed only on images themselves. to use insights that take a data-driven approach to design and decision making. A cognitive system is able to understand structured data and unstructured data, such as images, and can reason, learn and interact with people. runway fashion images and real-time social buzz and then used Watson Visual Recognition technology to not only analyse, but predict trends in colour. With a combination of IBM analytics and cognitive technologies, designers can gain new inspiration with access and insights gleaned from hundreds of thousands of social media images in an instant. Fashion designers usually start conceptualizing and designing products for the new season one year prior to the actual selling season (though in recent times this has been drastically reduced with the emergence of fast-fashion.). Therefore, for most apparel retailers (and fashion industry in general), knowing what the customers would like to wear next season is extremely important. A crucial aspect to achieve this is to have the ability of analyze current fashion trends and forecast future fashion trends. Complete the Look for fashion recommends apparels which would go well with the apparel the user is interested in. Complete the Look works on existing product text attributes in catalog and can handle category, color, pattern, brand, style, price, occasion, size etc. while utilizing the style rules.