Energy Compaction on Graphs for Motion-Adaptive Transforms
2015 (English)In: Data Compression Conference Proceedings, 2015, 457- p.Conference paper (Refereed)
It is well known that the Karhunen-Loeve Transform (KLT) diagonalizes the covariance matrix and gives the optimal energy compaction. Since the real covariance matrix may not be obtained in video compression, we consider a covariance model that can be constructed without extra cost. In this work, a covariance model based on a graph is considered for temporal transforms of videos. The relation between the covariance matrix and the Laplacian is studied. We obtain an explicit expression of the relation for tree graphs, where the trees are defined by motion information. The proposed graph-based covariance is a good model for motion-compensated image sequences. In terms of energy compaction, our graph-based covariance model has the potential to outperform the classical Laplacian-based signal analysis.
Place, publisher, year, edition, pages
2015. 457- p.
, Data Compression Conference, ISSN 1068-0314
Compaction, Covariance matrix, Forestry, Graphic methods, Image coding, Laplace transforms, Mathematical transformations, Matrix algebra, Principal component analysis, Trees (mathematics), Covariance modeling, Energy compaction, Image sequence, Karhunen Loeve Transform (KLT), Motion information, Motion-adaptive transform, Optimal energy, Temporal transforms, Data compression, Data Transmission, Energy, Mathematical Models
Computer Systems Probability Theory and Statistics
IdentifiersURN: urn:nbn:se:kth:diva-174701DOI: 10.1109/DCC.2015.86ISI: 000380409800064ScopusID: 2-s2.0-84938930606ISBN: 10.1109/DCC.2015.86OAI: oai:DiVA.org:kth-174701DiVA: diva2:868612
2015 Data Compression Conference, DCC 2015; Snowbird; United States; 7 April 2015 through 9 April 2015
QC 201511112015-11-112015-10-072016-08-23Bibliographically approved