Abstract: Photo Retrieval based on a Combination of Geo-referenced Attributes and Low-level Visual Features

This paper proposes a new efficient image retrieval method that automatically indexed for searching relevant images using a combination of geo-referenced attributes and low-level visual features. Photo image is labeled with its GPS coordinates and UTC date-time stamp at the moment of capture, and these data are then utilized to create two layer structures for spatial and temporal indexed for photo retrieval. In order to evaluate the performance of the proposed scheme, we assess the simulation performance in terms of average precision and F-score using digital photo collections. Comparing the proposed approach to search using visual content alone, an improvement of around 16~18% was observed in experimental trials. These results reveal that combining content and context information is markedly more effective and meaningful than using only visual content for this task.