Organizing Egocentric Videos for Daily Living Monitoring

Abstract

Egocentric videos are becoming popular since the possibility to observe the scene flow from the user's point of view (First Person Vision). Among the different assistive applications in this context there is the daily living monitoring of a user that is wearing the camera. In this paper we propose a system devoted to automatically organize videos acquired by the user over different days. By employing an unsupervised segmentation, each egocentric video is divided in chapters by considering the visual content. The video segments related to the different days are hence linked to produce graphs which are coherent with respect to the context in which the user acts. Experiments on two different datasets demonstrate the effectiveness of the proposed approach which outperforms the state of the art, both in accuracy and computational time with a good margin.