Machines are bringing an exciting edge to red carpet couture and daily wear. But can they displace humans?
In the 2006 film, The Devil Wears Prada, actor Meryl Streep who plays Miranda Priestly, a powerful fashion editor, gives her new assistant a dressing down for not understanding fashion. She tells her that fashion is whatever a select group of designers say it is. But what she fails to anticipate is how these czars of style will one day be challenged on their own turf by another set of fashionistas: machines.
As artificial intelligence (AI) pervades almost every field today, lines of an algorithm are now sashaying down the catwalk. India’s fashion and retail industry too have started to rely on the power of machines to come up with the latest styles.
“Fashion is like art, but there is also a lot of data involved in it,” says Myntra’s chief product officer Ambarish Kenghe. The country’s largest fashion e-tailer, in terms of sales, has been selling machine designed T-shirts under the brand name “Moda Rapido” and “Here and Now” for the past six months, with the products sometimes outselling other human augmented designs, according to Myntra.
The idea for “Moda Rapido” took shape during a session to spot trends across various sartorial details such as collars, plackets, sleeves and colours and how they can be used in combination with one another.
“We then took the next leap. We got the software to create various designs by using a combination of algorithms…and to find out which of these designs will practically work,” says Kenghe.
The first 30 machine made T-shirts were put on display without telling anyone how it was created.
“It worked better than human-augmented designs,” says Kenghe. “Now, we are expanding into 22 categories of clothing like kurtis and jeans.”
According to Kenghe, technology brings in efficiency and effectiveness. Something which a designer takes hours to create can be done with the help of a machine in minutes.
“Machine-generated or machine-augmented designs are selling two to three times more than the rest. The two brands are growing at over 200 per cent year-onyear,” says Kenghe.
“While expertise and forecasting agencies can be classified as one type of data, collective intelligence from millions of clients has a lot of merit,” says Stylumia founder-CEO Ganesh Subramanian, who worked closely with Myntra’s Moda Rapido software.
The AI-based SaaS startup in Bengaluru scans the internet for publicly available information about trends across categories and provides it to clients.
In fashion, the design process usually starts six months to a year in advance. “In the age of fast fashion, productivity can be improved 10 times when machines help with pattern recognition and sifting through large volumes of data,” says Subramanian.
“Our tool answers a question as simple as ‘what kind of products are working for H&M in UK right now?’ and you’ll get the answer in a few seconds,” he says.
“Clients say their sales are going up by 50 to 100 per cent by focussing on what sells best,” says Subramanian. The bootstrapped startup, which counts Wrangler India and Fastrack among its 40 clients, says machine learning will only play the role of enabler.
Meanwhile, innovations are on in couture to find out what is possible when fashion and technology come together on the red carpet. IBM’s Watson – an AI-based program – helped high-fashion label Marchesa create a garment for supermodel Karolina Kurkova to wear at the Met Gala.
In India, the tech giant roped in designers Gaurav Gupta and Falguni and Shane Peacock. In September, Gupta used cognitive technology to create a sari-gown that host Archie Punjabi wore at an award event. The dress lit up according to the personality of each award winner.
The software was able to study millions of images and information about a particular personality and give it a colour code. The design was about showing what is possible by bringing together technology and fashion,” says Gupta. #KhabarLive