Image-based product search to revolutionise online shopping


In July, e-commerce major Flipkart unveiled an image search feature on its mobile app that lets consumers get more apt search results while shopping online.

Via this feature, shoppers just need to click a picture or use a photo on their phones to find fashion products of same colour, pattern or style on Flipkart. Where Flipkart has added this technology as a feature on its mobile app, Flipkart’s rival Snapdeal has gone a step ahead with a view to build a best consumer experience and seller ecosystem.

Snapdeal recently announced a launch of visual-algorithm based micro site -- -- with the aim of enriching shopping experience online."The search methodology of is very detailed and customised to a consumer's preference to help them discover a product of choice with ease while saving time and effort," the company said in a statement.

Using Snapdeal’s, consumers will now be able to describe the product they wants to buy. Product description will be followed up by a show of the products that match the given description. Out of which, user can easily choose the most desirable option without having to browse through a diverse range of available products.

It also groups different types of products together using its image recognition software. After entering into one of the segments – striped shirts, colours, formal and informal ones -- the user can select an individual product(s).

Developed at Snapdeal's Multimedia Research Lab in Bengaluru, will act as a feedback system for the original website and will help the company understand more about consumer behaviour before the new technology in is integrated in the Snapdeal website.Commenting on the new feature, Anand Chandrasekaran, the chief product officer at Snapdeal, said: "We are a technology company with a focus on building cutting-edge services and products which make lives of our consumers simpler and convenient. is powered by advanced computer vision and machine learning algorithms.""The feature will enhance customer experience and make online shopping faster, better and more informative. With the introduction of, we are looking to create an experimental playground and set the base for our image modelling, product search and discovery technology,” Chandrasekaran added.

The idea is a brainchild of a tech startup called Fashiate that Snapdeal acquired back in March 2015. The new portal offers no separate buy option or payment integration. Buyers would be redirected to the Snapdeal website if they confirm a purchase.Computer vision, the newest technology in the Indian e-commerce space, is a growing discipline of computer science where computers are enabled with massive computing power to recognise images.

Traditionally, recommendations have worked with meta data tags which someone has to manually put in identifying different aspects of a clothing -- colour, pattern, sleeve, etc. The recommendation engine itself would look for similar clothes to the one being viewed based on the tags and then provide the results.

The other method has been collaborative filtering, which looks at the outfit the user is looking at and then says ‘Others who looked at this outfit, viewed these as well’. The visual search and recommendations technology analyses colour, pattern, gender, category of the clothing visually, and pulls up visually similar clothes to the one being viewed by the user. This also allows the company to generate tags automatically as opposed to needing someone to manually tag aspects of clothing.

Snapdeal said that it has plans to spend USD 100 million (Rs 640 crore) in the next three years on research and development as it looks to bring solutions to enrich online shopping experience. “We are focused on building in-house technology solutions. We plan to spend about USD 100 million on R&D in the next three years. This includes the Multimedia Research Lab in Bangalore,” said Chandrasekaran.

With Snapdeal investing heavily in new technologies related to multimedia and image in the coming quarters and Flipkart spending big on tech recruits, the budding multimedia searches category is expected to heat up in the e-commerce space.

Related Tags:, , ,

Hottest ProductsMore