Semantic Summarization of Egocentric Photo Stream Events

TL;DR

This paper presents a novel method for automatically summarizing egocentric photo streams using CNN-based filtering, relevance ranking, and a new evaluation metric, achieving high expert satisfaction.

Contribution

It introduces a new CNN-based filter, a relevance and novelty ranking process, and a novel evaluation metric for egocentric photo stream summarization.

Findings

95.74% expert satisfaction

Mean Opinion Score of 4.57/5.0

Effective reduction of redundancy and improved diversity

Abstract

With the rapid increase of users of wearable cameras in recent years and of the amount of data they produce, there is a strong need for automatic retrieval and summarization techniques. This work addresses the problem of automatically summarizing egocentric photo streams captured through a wearable camera by taking an image retrieval perspective. After removing non-informative images by a new CNN-based filter, images are ranked by relevance to ensure semantic diversity and finally re-ranked by a novelty criterion to reduce redundancy. To assess the results, a new evaluation metric is proposed which takes into account the non-uniqueness of the solution. Experimental results applied on a database of 7,110 images from 6 different subjects and evaluated by experts gave 95.74% of experts satisfaction and a Mean Opinion Score of 4.57 out of 5.0. Source code is available at https://github.com/imatge-upc/egocentric-2017-lta