Abstract
Thanks to the recent spread of smartphones, tablets and digital cameras, people can take photos easily anytime, and anywhere. We can estimate patterns of actions and movements of people by analyzing their photos treating them as a life log. Based on this discussion, we are developing a technique to recommend tourist spots based on the estimation of users' preferences for traveling plans from their past personal travel photos by using generic object recognition. This technique applies the genetic object recognition to a set of travel photos to extract preferred keywords of users' travel, and retrieves information using the keywords as query words. Also, the technique provides a visual user interface for structuring a large number of travel photos based on subject information and displaying a list of photos for each subject according to the purpose of travel. When a user selects photos of particular interest on the user interface, this user interface gathers keywords corresponding to the selected photos, extracts tourist spot information related to these keywords and recommends it to the user. This paper introduces the processing flow of the presented technique, and then a case study and evaluation with real travel photos.
