Estimation of arbitrary camera motion in MPEG videos

Ewerth, Ralph; Schwalb, M.; Tessmann, P.; Freisleben, Bernd

doi:10.1109/icpr.2004.1334181

Cited by 28 publications

(31 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…They achieved the best results for pan and tilt detection. Ewerth et al [1,3] obtained best results for zoom detection by computing the parameters of a 3D-camera model and applying thresholds for decision making.…”

Section: Related Workmentioning

confidence: 99%

Self-Supervised Learning for Robust Video Indexing

Ewerth

Freisleben

2006

2006 IEEE International Conference on Multimedia and Expo

View full text Add to dashboard Cite

The performance of video analysis and indexing algorithms strongly depends on the type, content and recording characteristics of the analyzed video. Current video indexing approaches often make use of thresholding techniques or supervised learning which requires labeling of possibly large training sets. Furthermore, the application of the same training model or parameters might lead to a suboptimal indexing accuracy for a given video. In this paper, we propose to use a novel self-supervised learning framework for robust video indexing to address this issue. Based on an initial classification result for a given video, the best features are selected by Adaboost and are then used to train SVM (support vector machine) classifiers, all on the given video. Finally, a specialized ensemble of classifiers is employed for the given video for decision making. Experimental results show that a state-of-the-art video cut detection approach can be significantly improved by the self-supervised learning approach.

show abstract

Section: Related Workmentioning

confidence: 99%

Self-Supervised Learning for Robust Video Indexing

Ewerth

Freisleben

2006

2006 IEEE International Conference on Multimedia and Expo

View full text Add to dashboard Cite

show abstract

“…It is applicable to the scenes able to be approximated by a planar surface for covering camera translations. A camera motion estimation technique for MPEG video is described in [4]. The camera translation parallel to the image plane is approximated by simply averaging the estimated rotation motion field.…”

Section: Introductionmentioning

confidence: 99%

Motion-Compensated Frame Prediction with Global Motion Estimation for Image Sequence Compression

Jiang

Dubois

2007

2007 Canadian Conference on Electrical and Computer Engineering

View full text Add to dashboard Cite

Based on a perspective projection model of sixparameter camera motions, a novel motion-compensated frame prediction approach is presented, taking into consideration all types of camera motions including camera translations. With the assumption of a continuously changing scene-depth distribution, a block-based scaled-depth estimation technique is proposed for obtaining the scaled-depth map of the predicted frame. The motion-compensated predicting frame is generated by using the global camera rotation and translation parameters combined with the scaled block-depth map. Compared with traditional block matching algorithms (BMA), more accurate predicting frames are obtained with the proposed motion-compensated frame prediction approach. As a result of fewer parameters required for motion compensation and reduced prediction residues, this approach is potentially more efficient if applied in frame prediction for image sequence compression. Experimental results on test images demonstrate the effectiveness of the proposed approach, and also demonstrate its superior performance over the traditional BMA applied for motioncompensated frame prediction.

show abstract

“…For different videos, this seems quite arbitrary as different strategies need to be applied [11,12].…”

Section: The Approachmentioning

confidence: 99%

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

et al. 2013

View full text Add to dashboard Cite

This version is available at https://strathprints.strath.ac.uk/48372/ Strathprints is designed to allow users to access the research output of the University of Strathclyde. Unless otherwise explicitly stated on the manuscript, Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Please check the manuscript for details of any other licences that may have been applied. You may not engage in further distribution of the material for any profitmaking activities or any commercial gain. You may freely distribute both the url (https://strathprints.strath.ac.uk/) and the content of this paper for research or private study, educational, or not-for-profit purposes without prior permission or charge.Any correspondence concerning this service should be sent to the Abstract: As an important feature in characterising 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 (SVM) for estimation in a regression model trained on a synthesised sequence. Experiments conducted on real sequences show that the proposed method yields much improved results in estimating camera motions whilst the difficulty in selecting valid macroblocks and motion vectors is skipped.

show abstract

Estimation of arbitrary camera motion in MPEG videos

Cited by 28 publications

References 8 publications

Self-Supervised Learning for Robust Video Indexing

Self-Supervised Learning for Robust Video Indexing

Motion-Compensated Frame Prediction with Global Motion Estimation for Image Sequence Compression

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

Contact Info

Product

Resources

About