Communications of the ACM
New Algorithms Locate Where a Video Was Filmed From Its Images and Sounds
A new system can geolocate videos by comparing their content against a global multimedia database.
Credit: Flickr-VV.AA
Ramon Llull University researchers have developed a system that can geolocate videos by comparing their audiovisual content with a worldwide multimedia database.
"In this field we use some physics and mathematical vectors taken from the field of recognition of acoustic sources, because they have already demonstrated positive results," says Ramon Llull researcher Xavier Sevillano.
The data was grouped in clusters so when using Ramon Llull-developed algorithms, they can be compared with those of a large collection of recorded videos already geolocated around the world.
The researchers used nearly 10,000 sequences as a reference from the MediaEval Placing task audiovisual database, a benchmarking initiative for assessment of algorithms for multimedia content. "The videos which are most similar in audiovisual terms to what we want to find are searched for in the database, to detect the most probable geographical coordinates," Sevillano says.
The system can locate 3 percent of videos within a 10-kilometer radius of their actual geographical location, and in 1 percent of cases it is accurate to one kilometer.
"This method could help rescue teams to track down where a person or group disappeared in a remote place, detecting the locations shown in the videos, which could have been uploaded to a social network before losing contact," Sevillano says.
From Plataforma SINC (Spain)
View Full Article
Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA
No entries found