Achievement

Trainees Emily Moxley and Jim Kleban (ECE) have covered significant ground with their SpiritTagger project. Geographically referenced photo data sets offer tantalizing potential for automated knowledge discovery in the world. By combining tag re-ranking based on geographic context with content-based image analysis SpiritTagger is able to suggest geographically relevant tags for photos newly tagged with GPS coordinates. These tag suggestions could help users organize their photo collections or improve retrieval systems. Our algorithm weights labels that correspond to pertinent objects, events, neighborhoods, and activities in a region. The results of SpiritTagger have yielded four papers. The project is significant because it’s the first use of geo-tagged photos (an emerging medium) for annotation. It opens up doors to innovative research in both spatial analysis, social media analysis, and annotation in cluttered areas. A demo is available at http://cortina.ece.ucsb.edu/index.php.