Abstract
With the prevalence of video surveillance and the extraordinary number of online video resources, the demand for effective and efficient content-based video analysis tools has shown significant growth in recent years. Human face has always been one of the most important interest points in automatic video analysis. In this paper, we designed a face-based video retrieval system. We analyzed the three key issues in constructing such systems: frame extraction based on face detection, key frame selection based on face tracking and relevant video retrieval using PCA-based face matching. In order to cope with the huge number of videos, we implemented a prototype system on the Hadoop distributed computing framework: DiFace. We populated the system with a baseline dataset consisting of TED talk fragments, provided by the 2014 Chinese national big data contest. Empirical experimental results showed the effectiveness of the system architecture and also the techniques employed.
Highlights
With the rapid development of smart city, security cameras and video surveillance has become ubiquitous in recent years [1]
We proposed a face-based video retrieval system
There exists three key issues in a face-based video retrieval system: 1) extraction of frames that containing human faces; 2) selection of key frames that are representative of a given video fragment; and 3) retrieval of relevant videos that contain the same person as in the given query video fragment
Summary
With the rapid development of smart city, security cameras and video surveillance has become ubiquitous in recent years [1]. Video sharing is one of the most popular features on online social networking websites [2] This results in unprecedented large number of video fragments, and the demand for effective and efficient automatic video content analysis tools has become urgent. Faces identified and extracted from video fragments can be used to develop content-based access and to facilitate intelligent multimedia tools and applications. We proposed a face-based video retrieval system. In order to cope with scalability, efficiency is a particular focus of designing solutions to these issues. We name the resulting distributed face-based video retrieval system DiFace. We review related work on automatic content-based video analysis with a specific focus on the human face angle.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have