Polar Rose Face Recognition Search Engine

Contributor: Anoop V.M.

The World Wide Web is increasingly becoming 'visually oriented', that is, photos and other rich media is replacing the text. About 10 million new photos are uploaded on a daily basis. For every eight to ten months the number is being doubled. This growing importance of photos in the web demands the need for an efficient method of managing online photos.

The search engines are the choice for managing the vast amount of textual data in the web. But no efficient method of search for photos is available even now. The 'image search' offered by famous search engines like Google, AltaVista etc is based on the nearby text analysis. This could be often misleading. The Sweden based company Polar Rose , which works on 'computer vision' technologies proposed a visionary new search based on the content in the photo.

The system could identify the faces in a photo. It then converts the 2D image to a 3D structure called 'face prints'. These face prints are used for the visual search applications. It also provides a tag for every face it identifies for the users to view and edit. Thus an user can get relevant information about a particular person on the photo from the tag provided with it. In general, it makes photos searchable by analyzing their content and recognizing the people in them.

The technology used here consists of two important parts. One is the unique algorithm which combines automatic face recognition with 3D modeling for visual search applications. This approach radically improves photo matching by compensating for variations in lighting, facial emotions and pose. The system also adopts a special method which makes use of the collective intelligence of the users who help train the software and tag names on people seen for the first time.

Scope of the topic:

In this era of growing importance of visual search ‘polar rose face recognition search engine’ is one of the major step towards it. The visionary and the sound technology will sure help a more meaningful search than the system used to day. The technology, in initial stage today, is likely to make a revolution in the field of online data search. In the beginning it will be available for the users as a browser plug-in.

There are a number of potential applications for Polar Rose's visionary search. For example, it enables efficient photo search as it is searching directly the content of the photo. The other simple examples, that the system provide in the near future are - search for more photos of the same person, collectively add information and tag people in online photos, automatically sort their online photos by the people appearing in them, to categories photos in an online photo album, to get additional information about any person photographed in a news site, in social networking sites, to identify people in an online conference. The uses of this technology in the field of ‘surveillance’ etc are on consideration.

The facial recognition search for other visual media such as video is the next development. The technology, which uses facial recognition and user-based contribution, is visionary enough to redefine the web of tomorrow. The out break of its news already brought concerns over ‘security’. Any way it could provide a more sensible and novel web experience with a lot of advantages and related controversies.