The way I see it, the web is increasingly becoming visually oriented, progressing from text to photos and other rich media such as video and podcast. Polar Rose claims that close to 10 million new photos are uploaded on a daily basis, a number which is doubling every eight to ten months. We should thanks to the digital camera innovation and internet for the exponential growth. Perhaps a picture does tell a thousand words.
Even so, most of the photo are unreachable due to lack of algorithm that could surface them like the way text search does. Since the day of Altavista, Inktomi, and Google we haven't seen any new revolutionary of photo search technology without the "human" intervention.
To overcome this issue, Polar Rose technology makes photos searchable by analyzing their content and recognizing the people in them. Yeah it's another face recognition technology just like Riya that I've discovered earlier. The Polar Rose's face recognition technology used was originally developed by CTO Jan Erik Solem during his M.Sc and Ph.D. stints at the universities of Lund and Malmö in southern Sweden. It's unique in that they are able to extract 3D information from regular 2D images.
However "human" users are still needed as to help to train the application recognize each individual in the photo. Polar Rose offers a plugin to be installed into your Firefox or Internet Explorer browser. The plug-in detects people in online photos and places their signature rose approximately where the pinhole of their shirt would be. A click on the rose will bring up a tooltip with relevant information, including name and other photos found of the same person. You can apply to become a beta tester but the plugin will be released only during the first quarter of 2007.
Face recognition is cool but like I said earlier in my previous post, if this technology going to be a public tools it does raise some privacy issues. To avoid this, perhaps it can be tune by identify object or item instead.
No comments:
Post a Comment