|IARPA Awards $15.6 Million To Find Where A Photo Was Taken|
|Written by Alex Armstrong|
|Sunday, 29 July 2012|
You take lots of photos and of course you know where they were taken but the intelligence services would like a system than can scan in a photo and tell you exactly where it was taken - anywhere on earth.
We reported on the proposed US Intelligence Advanced Research Projects Activity (IARPA), Incisive Analysis Office project to identify where a photo was taken last year. The news now is that it has found a company, ObjectVideo, to take the project on. The award is one of four made to companies. The Finder project aims to identify the geolocation of a photo or still from a video using only publicly available data.
The system can accept help from an analyst who will probably be able to provides some hints on roughly where the photo was taken.
The Finder Program aims to build on existing research systems to develop technology that augments the analyst’s abilities to address the geolocation task. Technical innovations of the system will include the integration of analysts’ abilities and automated geolocation technologies to solve geolocation problems; the fusion of diverse, publicly-available, but often imperfect data sources; and expansion of current automated geolocation technologies to work more efficiently and accurately over all terrains and large search areas.
If successful, the Finder system will deliver a rigorously-tested technology solution for image and video geolocation tasks in any outdoor terrestrial location.
It makes you wonder if there is enough information contained in a typical photo or video to pin it down to even a rough geographical location. Having the time and date the photo was taken might help. Recognizing geological features and local topology also seems like a good approach.
Given that the award is $15.6 million, which seems a lot, phase one of project is estimated to be completed in 2014, which is hardly a tight deadline. It really does sound like the sort of project that might turn up on KickStarter or something that might be done just for fun. Could it be built into the next generation of cameras?
Of course, as mentioned in the first news report, as soon as an automatic image geolocation system is perfected we can start work on a system to spoof it.
or email your comment to: firstname.lastname@example.org
To be informed about new articles on I Programmer, install the I Programmer Toolbar, subscribe to the RSS feed, follow us on, Twitter, Facebook, Google+ or Linkedin, or sign up for our weekly newsletter.
|Last Updated ( Sunday, 28 February 2016 )|