Leveraging Deep Learning to Improve the Retail Experience

leveraging-deep-learning-to-improve-the-retail-experience

During the dot-com boom, online clothing sales were predicted to grow to 40% -50% of total sales. Although online sales of some other kinds of merchandise, such as books, have reached 50% of the market in the past 15 years, the percentage of online clothing sales hovers around 20%. The difficulty in finding the correct size and fit is one of the primary reasons that consumers are reluctant to buy clothes online. And their concern is not groundless; sizing varies among clothing manufacturers, and it is difficult to ascertain fit from online images.  Consequently, 30%-40% of online clothing purchases are returned. In addition, certain kinds of clothes fitting, such as for shoes and bras, are difficult even in brick-and-mortar stores.

Computer vision has been used to address the problem of finding the right fit online by using mobile devices, digital scanning, and deep learning. For example, San Francisco-based ThirdLove offers a free app for iPhones that enables women to find the right bra from home. The app calculates a woman’s measurements, cup size, shape, and fit by comparing the distance between body contours in front and side view photos that the woman takes while wearing a supportive bra and fitted top, standing before a mirror, and holding her iPhone camera at the level of her navel. The app bases the sizing on instantaneous calculations, and protects consumer privacy by not storing the images on the cloud or the iPhone.

The Stockholm-based company Volumental offers computer vision applications for sizing shoes and eyewear. Its VANDRA hardware platform contains four depth cameras that make a 3D, volumetric scan of a foot in 2 seconds, including areas that retailers normally do not measure. The data is sent to a tablet, which a salesperson consults in order to make shoe recommendations from store inventory or other retailers. As another application of computer vision fitting, Volumental’s VACKER platform uses 3D face scanning to assist opticians and customers to find the most flattering eyewear for different face shapes and sizes.  While its technology is currently store-based, Volumental anticipates that depth cameras will be eventually be integrated into smartphones, tablets, and laptops in 2017, allowing customers to use VANDRA and VACKER for online purchasing as well.

Technologies that accurately assess clothing/accessory size and fit provide a more satisfying and individualized retail experience for customers, not to mention enhancing their overall comfort by ensuring that they wear clothes that fit and flatter them. At the same time, computer vision fitting cuts down on retailers’ expenses related to product returns and customer complaints.  By providing more seamless online shopping, computer vision fitting also supports online markets, expands customer choice, and will eventually provide access to personally tailored clothes.

According to Tractica’s research, companies like ThirdLove and Volumental make retail one of the industry sectors best positioned to leverage artificial intelligence (AI) and deep learning. In our Deep Learning Intelligence for Enterprise Applications report, we forecast that spending on AI software in the retail sector will grow from $6 million in 2015 to $568 million by 2024.

DLE-15 Retail chart

 

Comments are closed.