Video matching using DC-image and local features

Bekhet, Saddam and Ahmed, Amr and Hunter, Andrew (2013) Video matching using DC-image and local features. In: 2013 World Congress on Engineering, WCE 2013, 3-5 July 2013, London, UK.

Full content URL: http://www.iaeng.org/publication/WCE2013/WCE2013_p...

Documents
Video matching using DC-image and local features
[img]
[Download]
[img]
Preview
PDF
__ddat02_staffhome_jpartridge_WCE2013_pp2209-2214.pdf - Whole Document

719kB
Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive

Abstract

This paper presents a suggested framework for video matching based on local features extracted from the DCimage of MPEG compressed videos, without decompression. The relevant arguments and supporting evidences are discussed for developing video similarity techniques that works directly on compressed videos, without decompression, and especially utilising small size images. Two experiments are carried to support the above. The first is comparing between the DC-image and I-frame, in terms of matching performance and the corresponding computation complexity. The second experiment compares between using local features and global features in video matching, especially in the compressed domain and with the small size images. The results confirmed that the use of DC-image, despite its highly reduced size, is promising as it produces at least similar (if not better) matching precision, compared to the full I-frame. Also, using SIFT, as a local feature, outperforms precision of most of the standard global features. On the other hand, its computation complexity is relatively higher, but it is still within the realtime margin. There are also various optimisations that can be done to improve this computation complexity.

Keywords:Compressed domain, Dc-image, Sift, Video matching, Video Similarity
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
Divisions:College of Science > School of Computer Science
ID Code:12680
Deposited On:13 Dec 2013 11:11

Repository Staff Only: item control page