Research Article
Egocentric Video Summarization Based on
People Interaction Using Deep Learning
Humaira A. Ghafoor,
1
Ali Javed ,
1
Aun Irtaza,
2
Hassan Dawood,
1
Hussain Dawood,
3
and Ameen Banjar
3
1
Department of Sofware Engineering, University of Engineering and Technology, Taxila 47050, Pakistan
2
Department of Computer Science, University of Engineering and Technology, Taxila 47050, Pakistan
3
Faculty of Computing and Information Technology, University of Jeddah, Saudi Arabia
Correspondence should be addressed to Ali Javed; ali.javed@uettaxila.edu.pk
Received 26 July 2018; Accepted 18 November 2018; Published 29 November 2018
Academic Editor: Stanislav V´ ıtek
Copyright © 2018 Humaira A. Ghafoor et al. Tis is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
Te availability of wearable cameras in the consumer market has motivated the users to record their daily life activities and post
them on the social media. Tis exponential growth of egocentric videos demand to develop automated techniques to efectively
summarizes the frst-person video data. Egocentric videos are commonly used to record lifelogs these days due to the availability of
low cost wearable cameras. However, egocentric videos are challenging to process due to the fact that placement of camera results
in a video which presents great deal of variation in object appearance, illumination conditions, and movement. Tis paper presents
an egocentric video summarization framework based on detecting important people in the video. Te proposed method generates
a compact summary of egocentric videos that contains information of the people whom the camera wearer interacts with. Our pro-
posed approach focuses on identifying the interaction of camera wearer with important people. We have used AlexNet convolutional
neural network to flter the key-frames (frames where camera wearer interacts closely with the people). We used fve convolutional
layers and two completely connected hidden layers and an output layer. Dropout regularization method is used to reduce the
overftting problem in completely connected layers. Performance of the proposed method is evaluated on UT Ego standard dataset.
Experimental results signify the efectiveness of the proposed method in terms of summarizing the egocentric videos.
1. Introduction
Te introduction of wearable cameras in 1990s by Steve Mann
has revolutionized the IT industry and created a deep impact
in our daily lives. Te availability of low cost wearable cameras
and social media has resulted in an exponential growth of
the video content generated by the users on daily basis. Te
management of such a massive video content is a challenging
task. Moreover, much of the video content recorded by the
camera wearer is redundant. For example, narrative clip and
GoPro cameras record a large amount of unconstrained video
that contains much of the insignifcant/redundant events
beside the signifcant events. Terefore, video summarization
methods [1, 2] have been proposed to address the issues asso-
ciated with handling such a massive and redundant content.
Egocentric videos are more challenging to address for
summarization due to the presence of jitter efects experi-
enced because of camera wearer’s movement. Accurate fea-
ture tracking, uniform sampling, and broad streaming data
with very refned boundaries are the additional challenges to
lifelogging video summarization. To address the aforemen-
tioned challenges associated with the egocentric videos, there
exists a need to propose efective and efcient methods to
generate the summary of full-length lifelogging videos. Some
distinctive egocentric video recording gadgets are shown in
Figure 1. Te focus of these egocentric video recordings is on
activities, social interaction, and user’s interests. Te objective
of the proposed research work is to exploit these properties
for summarization of egocentric videos. Egocentric video
summarization has useful applications in many domains, i.e.,
Hindawi
Mathematical Problems in Engineering
Volume 2018, Article ID 7586417, 12 pages
https://doi.org/10.1155/2018/7586417