7 May 2013 Reliable camera motion estimation from compressed MPEG videos using machine learning approach
Zheng Wang, Jinchang Ren, Yubin Wang, Meijun Sun, Jianmin Jiang
Author Affiliations +
Abstract
As an important feature in characterizing video content, camera motion has been widely applied in various multimedia and computer vision applications. A novel method for fast and reliable estimation of camera motion from MPEG videos is proposed, using support vector machine for estimation in a regression model trained on a synthesized sequence. Experiments conducted on real sequences show that the proposed method yields much improved results in estimating camera motions while the difficulty in selecting valid macroblocks and motion vectors is skipped.
© 2013 Society of Photo-Optical Instrumentation Engineers (SPIE) 0091-3286/2013/$25.00 © 2013 SPIE
Zheng Wang, Jinchang Ren, Yubin Wang, Meijun Sun, and Jianmin Jiang "Reliable camera motion estimation from compressed MPEG videos using machine learning approach," Optical Engineering 52(5), 057401 (7 May 2013). https://doi.org/10.1117/1.OE.52.5.057401
Published: 7 May 2013
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Cameras

Video

Motion estimation

Video compression

Video surveillance

Motion models

Video processing

RELATED CONTENT

Mosaics from MPEG-2 video
Proceedings of SPIE (July 01 2003)
NeTra-V: toward an object-based video representation
Proceedings of SPIE (December 23 1997)
Compressed video indexing based on object motion
Proceedings of SPIE (May 30 2000)
Mosaicing of MPEG compressed video: a unified approach
Proceedings of SPIE (November 02 2004)

Back to Top