Reliable camera motion estimation from compressed MPEG videos using machine learning approach

Wang, Zheng and Ren, Jinchang and Wang, Yubin and Sun, Meijun and Jiang, Jianmin (2013) Reliable camera motion estimation from compressed MPEG videos using machine learning approach. Optical Engineering, 52 (5). 057401. ISSN 0091-3286 (https://doi.org/10.1117/1.OE.52.5.057401)

[thumbnail of CME-OEL-ver6d]
Preview
PDF. Filename: CME_OEL_ver6d.pdf
Preprint

Download (873kB)| Preview

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.