Visual Search Engine Debuts

Search company Riya, which has been developing facial recognition technology for locating individuals in photographs, has spun off its image search technology onto a site called The service enables users to search for images by appearance - not just text keywords.

Initially, is focusing on shopping categories -- handbags, jewelry, shoes, and watches -- and is adding items at a rate of 30,000 per day. Users can search using descriptive terms such as a color or feature of a product, or even upload a photo to find similar items.

"What this means - If you see an image on the web, like a watch that Paris Hilton is wearing in the picture to the left, and use it as an image query, will return results showing watches that look very similar," explained Michael Arrington on the TechCrunch blog. "If you enter a text query, like 'brown boots pointed toe,' will convert that query into variables in the visual signature and look for related image results."


Riya says image search accounts for eight percent of all Web searches, but current services simply utilize keywords attached to photos. Google has even resorted to recruiting individuals to help add tags to its images index.

It's not clear if and when will move beyond shopping, but Riya feels the space is big enough to allow quite a bit of growth for the company.

" is the first major new innovation to come to the world of image search in the last five years. Now, Web searchers will no longer have to rely on the tags that may or may not be attached to a given photo or product image," said Riya CEO and co-founder, Munjal Shah.

5 Responses to Visual Search Engine Debuts

© 1998-2022 BetaNews, Inc. All Rights Reserved. Privacy Policy - Cookie Policy.