Hot Photo-Spot Annotation using photo attributes and image features

Abstract

In our previous work, we presented a method of extracting a pair of a major photo-spot and its hot-period, which is called a hot photo-spot, from a large number of geotagged photographs with times- tamps that many people have taken. However, as for explaining the hot photo-spots extracted, it was in general difficult to annotate them clearly since each of them can have a variety of photos. In this paper, we propose a method of explaining each hot photo-spot by classifying the photos in it based on their im- age features and attributes such as their geotags and timestamps. Using real data from "Flicer data", we experimentally demonstrate the effectiveness of the proposed method.