Video segmentation of life-logging videos

Abstract

Life-logging devices are characterized by easily collecting huge amount of images. One of the challenges of lifelogging is how to organize the big amount of image data acquired in semantically meaningful segments. In this paper, we propose an energy-based approach for motion-based event segmentation of life-logging sequences of low temporal resolution. The segmentation is reached integrating different kind of image features and classifiers into a graph-cut framework to assure consistent sequence treatment. The results show that the proposed method is promising to create summaries of everyday person's life. © 2014 Springer International Publishing.