Preference-Aware Video Summarization for Virtual Tour Experience

Recently, there are increasing demands for tour recommendations using videos because more tourists search and watch tourism videos when planning a sightseeing tour. However, searching desired videos are very hard (even not possible) or require significant labor. In addition, since each searched video is on a fixed route, it is necessary to search other route videos when it does not match the user's desire. In this paper, we propose a system for curating a tour video adjusted to the intention or preference of each individual user from consumer generated media (CGM) including videos, photos and comments posted on each sightseeing spot/route through Social Networking Service. This system creates a tour route based on user's preference and a summarized video along the route. For creation of a summarized video, we propose a video summarization algorithm that maximizes viewing satisfaction within a specified playback time. Users can plan favorite a tour easily by watching this summarized video. In order to evaluate effectiveness of our method, we compared watching a summarized video with searching by Internet. As a result, we found that watching a summarized video can help a user imagine a tour easily, improve sightseeing and increase his/her satisfaction.