Title :
Multi-frame motion estimation: application to motion compensated prediction
Author :
Gibson, David ; Spann, Michael
Author_Institution :
Birmingham Univ., UK
Abstract :
Motion trajectories provide a localised description of motion over multiple frames. An approach is developed for the estimation of a spatially distributed set of such trajectories with the aim to adequately describe all of the motions present within an image. Spatial smoothness of the motion field is enforced in a Markov Random Field (MRF) framework, with a coupled line-process field to handle motion boundaries. Use is made of the resultant set of motion trajectories for the purpose of motion compensated prediction
Keywords :
Bayes methods; Markov processes; covariance analysis; data compression; image sequences; interpolation; motion compensation; motion estimation; video coding; Bayesian framework; Markov random field framework; coupled line-process field; covariance match; feature tracking; image sequences; interpolation; motion boundaries; motion compensated prediction; motion model; motion trajectories; multi-frame motion estimation; polynomial model; spatial smoothness; spatially distributed; video compression; Acceleration; Application software; Computer vision; Markov random fields; Motion estimation; Smoothing methods; Software libraries; Tracking; Trajectory; Video compression;
Conference_Titel :
Circuits and Systems, 1999. ISCAS '99. Proceedings of the 1999 IEEE International Symposium on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-5471-0
DOI :
10.1109/ISCAS.1999.779941
