India's Myntra launches AI-driven feature for customised fashion tips

My Stylist utilises AI to offer customers complete shoppable outfit recommendations based on four different factors—uploaded images from their offline wardrobes, their purchase history on the app, browsing history on the app, and what’s currently trending on the platform, thus giving instant access to styling options and enhancing their shopping experience, making it visual, fun, and instantly gratifying, Myntra said in a press release.

The feature throws recommendations from around 4.5 lakh styles from the fashion and lifestyle categories currently, which includes all topwear, bottomwear, footwear, and accessories. My Stylist is currently the highest used feature on the floating action button or M-Explore on the homepage of the app, with a high engagement rate and click through rate (CTR) of over 65 per cent. The company expects nearly about 80 per cent of its customers to engage with recommendations from the My Stylist feature on trial by the end of this year.

For customers who have shopped before on Myntra, My Stylist helps them find and shop for a complete look based on any apparel product they bought recently on the app, recently browsed products, or popular outfits on the platform based on which outfits are generating maximum engagement from users. For new customers, the feature also enables them to upload images of any product from their offline wardrobe, and get pairing ideas. New and existing users can also play around with the feature by taking a picture or uploading an image from their phone gallery of any offline product from their wardrobe and checking out pairing ideas to complete an outfit. The feature lets users ‘shop the entire look’ at the click of a button.

Myntra’s in-house AI technology, machine learning models, and computer vision algorithms, along with expertise of subject matter experts in curation and styling, are used to power My Stylist’s Fashion Object Detection, Image Search, and Outfit Recommendations features. These aspects use an AI algorithm that taps into the platform’s catalogue data via deep convolutional neural networks and is also trained using a Bi-LTSM model that expands dataset from Myntra’s catalogue, which enables the feature to handle a high-volume of requests.

“The launch of My Stylist takes our commitment of democratising fashion using technology, a notch higher. Powered by machine learning and AI-based technology, this first-of-its kind innovation at scale in the fashion space in India, is poised to further take the customers’ shopping experience to the next level by recommending looks with an understanding of fashion. Our team of experts has worked tirelessly to ensure that this feature is user-friendly, and intuitive, and from what we have gathered since the launch, we are confident that our customers all across will love to have access to this expert stylist at their fingertips through the much-loved Myntra app,” said Raghu Krishnananda, chief product and technology officer, Myntra.

The feature currently covers the fashion and lifestyle segments, with plans in the pipeline to include other categories including beauty and an increased number of styles as well. Myntra also plans to include elements like Your Digital Wardrobe and Save the Look, along with additional customisable user options, which consider factors like trends, age, gender, body type, and personal style, in subsequent phases.

Fibre2Fashion News Desk (NB)