Researchers at the University of Cambridge have developed at system called 'PoseNet' that can recognize the location of any photo without GPS. While many people use GPS to tag the location of their photos, this system offers several advantages that make it superior to GPS-based technology.
The PoseNet system is designed to instantly recognize the location of any photo within six feet of distance. The system relies on convolutional neural networks to process visual information and pick up different details such as faces. The system is superior to GPS and other types of image recognition technologies because it does not require an existing infrastructure, it gives an estimate of orientation and it is significantly more accurate even when indoors. In addition, the system is extremely fast and requires very little storage data.
While PoseNet is currently only available as part of a tech demo, its many advantages mean that it could one day replace current image recognition technologies.
GPS-Free Location Tags
More Stats +/-
Photo App-Based Auctions
Geo-Tagged Confession Apps
Crowdsourced Location Tagging
Creepy Geo-Tagging Applications
Augmented Reality Maps