Recently, we launched the detection of duplicate faces. Since we rely only on the face, we are able to match any version of the image as soon as the face remains the same. This is of interest not only to find the best quality version of an image but also to discover all kinds of modifications.
Only a few days after moving from private to public beta we’ve now opened up for partners as well by releasing the Polar Rose Widget.
This first release is initially for partners to test the functionality on their websites. Over the next few weeks will be getting to feedback and making updates and improvements. We will announce when the widget is fully open to all.
Jalbum is one of the fastest growing photo sharing communities in the world. Jalbum photo albums can have any look, functionality and can be published to any site. Millions of photo albums are distributed over 100’s of thousands of web servers globally, with a current rate of 90 million monthly page views.
Our friends at Jalbum are currently finalizing the remaining tasks to have the widget available to all their users as an opt-in feature. You can get a sneak preview of the widget in action at http://davidekholm.jalbum.net/faces/.
We’re very happy to have Jalbum as our first partner and welcome the Jalbum team and users to Polar Rose.
PS. Read more about the partner widget here http://info.polarrose.com/partners/ where you’ll find a presentation, technical specifications and everything else needed to get started. Remember that we’re some days off from the public release so changes can still occur and please use for testing only.
At Polar Rose we use computer vision to analyze the content of online photos and recognize the people in them.
Recognizing people in photos is a great challenge because the computer must interpret the photos to extract information above a pure pixel-level, something we humans do easily every day. The main problems when working with general photos are changes in lighting and pose which have great effect on the appearance of a face in a photo.
To cope with this our software creates a virtual 3D portrait of a person so that we can factor out and compensate for these effects. The technology is based on computer vision research for computing 3D data using ordinary photos. Statistical relationships between what can be observed in the 2D photos and how this relates to the possible 3D shape variations gives us a way to get 3D using a single picture. Even if this 3D portrait is just an approximation and not a measurement it is a very good one which adds significant boost to recognition.
Here’s a screenshot of our good friend Ola Ahlvarsson as his 2D picture is converted into a 3D portrait. View the entire screen recording here.
