High Time for Machine Learning: Artificial Intelligence and fashion cross paths

Falguni and Shane are now using Machine learning platform of IBM, Watson to discover the possibilities in fashion. This is where Artificial Intelligence and Fashion is crossing paths. 

Mumbai-based fashion designers Falguni and Shane Peacock have collaborated with Watson, creating designs that were showcased at the Grand Hyatt Mumbai. Sriram Raghavan, director of IBM Research India, unpacks the process: “The way it works is that we teach Watson (Machine Learning Platform of IBM) to understand fashion blogs, reviews, colour, silhouettes. We build an ontology of fashion language, like what does A-line mean, what does mandarin neck mean, infusing it with a deep level of expertise.” Once “super charged”, it works like a fashion expert, providing research as well as suggestions and visual representations.

Machine Learning in Fashion

Machine learning algorithm was fed over 600,000 images from fashion weeks that have been held over the last decade in Milan, New York, Paris and Tokyo, plus Bollywood data from the last four decades. Then Falguni and Shane worked with the tool to design a line called the Future Of Bollywood Fashion. “Let’s say you ask from algorithm about the little black dress. It will respond with images of the dress, along with when and where it was shown,” says Shane.

Machine learning’s contribution in other industries seems a lot more fascinating. For instance, the Memorial Sloan Kettering Cancer Center in New York has been using Watson to “augment” cancer care since 2013. And in a recent project, Watson was used by students of the Maharashtra Institute of Technology, Pune, to create Chintu, a robot that assists senior citizens in everyday chores. The role of AI in these cases reveals an immediate connect. In the whimsical world of fashion, however, Watson’s nascent intelligence could use some imagination.

Learn Machine Learning from Industrial Experts: Click Here

Don't miss out!
Subscribe To Our Newsletter

Learn new things. Get an article everyday.

Invalid email address
Give it a try. You can unsubscribe at any time.