• Home
•  
• Call for Papers
• Paper Status
• Registration
• Workshops
• Exhibits
•  
• Technical Program
• Side Meetings
• Tutorials
• Plenaries
• Special Sessions
• Panel Sessions
• My Schedule
•  
• Practicalities
• Housing & Venue
• Special Events
• Visits
• Supporters
•  
• Contact / Discussion
• Organizing Committee
• Program Committee
• Review Committee
•  
• Related Conferences

Technical Program

Paper Detail

Paper:	IMDSP-P1.12
Session:	Motion/Disparity Detection and Estimation
Time:	Tuesday, May 16, 10:30 - 12:30
Presentation:	Poster
Topic:	Image and Multidimensional Signal Processing: Motion Detection and Estimation
Title:	Robust Global Motion Estimation from MPEG Streams with a Gradient Based Refinement
Authors:	David Corrigan, Anil Kokaram, University of Dublin / Trinity College, Ireland; Renan Coudray, Bernard Besserer, University of La Rochelle, France
Abstract:	In this paper a new robust algorithm for Global Motion Estimation from an MPEG stream is proposed. The approach uses a hybrid of two Global Motion Estimation algorithms to improve the accuracy and robustness of the estimate. The first method is a non-parametric technique which uses the block based motion vectors obtained from an MPEG2 stream and also generates a coarse background segmentation of the frame. The second is a gradient based technique which estimates the parameters from the image data. This hybrid approach shows significant reduction in mean compensation error over the non-parametric method. The algorithm is then applied to mosaicking in sports clips in which global motion estimates are used to make a panorama of a clip.