Some favorite site feeds aggregated locally: iPhone Development RSS   Adobe Labs RSS   Macrumors RSS

Monday, September 22, 2008

Facial recognition

Monday, September 22, 2008   

Blue Lava has announced that its iLovePhotos application has moved into a public beta. As photos are imported, iLovePhotos automatically detects individuals' faces. 

The user can then add tags to the faces, and in future imports, people will be recognized and properly tagged automatically. After the process is complete, individuals can choose to filter their library to only show photos of specific people. 

The software further uses the tags to pull information from Address Book, relying on this to set up automatic sharing feeds.

Alongside the public beta, a new Flash-based version is being introduced. Users can access the online software from any Mac or PC without downloading any software, and import their photos straight from Flickr.

**I haven't played with this at all and assume that the facial recognition is being done on the server side... but I suppose the technology is available to the Flash Player to whip something like that up -- after a whole lot of math and hard work. A Flash interface (which this probably is) for anything though is cool.
 
 Return to the main page
Comments:

There are currently 2 Comments:

Anonymous Paul Mayne said...
“It detects faces, but I don't see anything about it being able to auto tag faces. So no real person detection. (based on watching the demo)”
 
Anonymous Cory said...
September 23, 2008 12:59 AM
“Thanks a lot for writing about us. To clarify, iLovePhotos finds the faces in your photos and provides a streamlined tagging interface that separates the faces from the photos so you can sort out who's who really quickly.

Organizing your photos by people allows you to automatically share your photos with the people in them...so just by tagging a person in one of your photos, that photo can be uploaded to iLovePhotos.com, and an email sent to the person in that photo with a link.

We're currently working on getting facial recognition into the software and website, but as a first step, we've completely streamlined the tagging process.

I'd love to hear about any feedback you may have using the service.

Thanks!
Cory”
 
 Leave a comment