This is a bio-inspired image recognition business spun out of Imperial College London, in February 2009. After six and a half years of research into “human visual cortex”, to understand how humans see – and two years building algorithms and computer models which can accurately mimick human visual recognition – Cortexica is now going live. It’s Visual Search API is now in closed beta, which developers can request access to by emailing api@cortexica.com
One of the first uses of Cortexica’s image system will be seen in the (now live) WINEfindr iPhone app, which claims to be the “world’s first visual search enabled price comparison wine app”. Essentially, users can take a picture of a bottle of wine and WINEfindr returns multiple results so the prices of different retailers can be compared and a whole host of data.
[Download PP presentation here]
Founder and CEO Steve Semanzato said in his pitch to GeeknRolla’s first start-up panel of the day: “Where visual search has been in the past, we have taken pictures with phones but now we’re starting to see just how our phones can relate to the label on a product… and we want to expand the horizons of what you can do with this.”
Through Cortexica’s API, app-makers and online businesses can add visual search images to their mobile product. “We’re finding a whole host of people are coming to us – ranging from grocery products, wine sellers,” says Semanzato.
The company says its monitoring service can theoretically index 200 million frames of video per day and it’s offering that capability on a commercial basis to media monitoring agencies.