It seems that the image based product search is the new trend for eCommerce portals in India. Earlier, Flipkart implemented image-based search for its fashion category and now, two of India’s fastest growing eCommerce startups for women fashion – CraftsVilla and Voonik are rolling out this feature soon.
Through images based search, you can take a picture of a product you are interested in and the portal will show similar products to you. Do note though, that this tech has already been implemented by numerous app-based startups, like Roposo. We had an entry in one of our Pitch-Off events as well, for an app named SnapShopr using a similar image recognition tech to help you find your favourite apparel.
Both Voonik and CraftsVilla are going to include their entire portfolio. Thus, instead of a text-based search, you can find products via images.
As per Sujayath Ali, CEO & co-founder of Voonik, visual search for fashion provides an advantage. Manoj Gupta, co-founder & CEO at Craftsvilla, said,
“If the customer likes a colour or pattern, they’d want to be shown a similar kind of product and this feature is incredibly useful.”
Until now, eCommerce portals were using text-based search. The products are tagged with keywords and so are the product images. Whenever a user enters a keyword, it searches for the particular keyword and shows products with that keyword, pretty much how you search on our website as well.
The introduction of this image-based search feature will make it easier for customers to look for a particular product. However, the technology used to develop the feature is not that easy.
To recognize images and giving output of similar products, programmers use computer vision, which requires a great amount of computing power to recognize images. While MADStreetDen is using its ‘Visual Search’ product for implementing this, it is also using one of its other products, called the MADstack. This is a cloud based platform that offers several AI and Computer Vision modules, just an API call away. The API includes platform for several functions like Object recognition, gaze tracking, emotion-expression detection, head and facial gestures, 3D facial reconstruction etc., to give a more visually dynamic experience to consumers.
Microsoft’s PixelSense is pretty much an example of Computer Vision Tech, though it uses physical interaction as a medium to recognise the input.
The computer vision technology for image recognition to CraftsVilla and Voonik will be provided by Mad Street Den, a Chennai based start-up which focuses on computer vision. Founded in 2013, Mad Street Den analyses an image for its color, pattern and other attributes visually, and pulls up similar products. It received $1.5 million seed funding this year.
Ashwini Asokan, co-founder of Mad Street Den, said,
“We’re not just a visual search company. We’re an artificial intelligence company that has a horizontal computer vision platform going across recognition of objects, gestures and emotions.”
CraftsVilla was founded in 2011 by Bhavik Jhaveri, Monica Gupta, Sarvajeet Chandra, Tirath Kamdar and Manoj Gupta. It has raised about $20.5 million in 3 funding rounds from Sequoia Capital, Lightspeed Venture Partners and Nexus Venture Partners.
Voonik was founded in 2013 by Navaneetha Krishnan and Sujayath Ali to help people dress well within their budget, by offering style advice and hand-selected recommendations, based on the customer’s style preferences and body type. Till now, it has raised $5.5 million from SeedFund and Sequoia Capital.