Visual shopping search on mobile phones

 

This is the buzz of the day: a company named Evolution Robotics will launch a visual search technology for the iPhone in June:

The idea is quite simple, yet difficult to implement: using the embedded camera on the iPhone (and potentially any other camera-enabled mobile phone), you take a picture of a product (CD, Book, Movie) and send it to the Evolution Robotics’ servers; in return, you get a mail with all details about this product: reviews, videos…; and the option to buy the item on iTunes.

The product of Evolution Robotics – called ViPR – seems pretty advanced and efficient according to the its website:

  • [...] Works for a wide range of viewing angles, lens distortion, imager noise, and lighting conditions
  • Works even when a large section (up to 90%) of the pattern is occluded from the view by another object
  • Can simultaneously recognize multiple objects
  • Can handle databases with thousands of visual patterns without a significant increase in computational requirements (the computation scales logarithmically with the number of patterns) [...]
  • ViPR technology can be purchased for development [...] It can also be licensed as a standalone technology to be included in your products.

The technology looks amazing, yet some more pragmatic solutions already exists, like the recognition of bar codes. But taking pictures of the product is much more intuitive, and beyond the “exact match”, we could imagine bringing similar products as results (think fashion for instance).

I wonder the cost to manage a corpus of visual patterns; but it could be really interesting to plug this engine on a mobile comparison engine like Kelkoo Mobile…

April 23, 2008

• •

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading ... Loading ...