Locally oriented optical flow computation

Simple item page
dc.contributor.authorNiu, Y.
dc.contributor.authorDick, A.
dc.contributor.authorBrooks, M.
dc.date.issued2012
dc.description.abstractThis paper proposes the use of an adaptive locally oriented coordinate frame when calculating an optical flow field. The coordinate frame is aligned with the least curvature direction in a local window about each pixel. This has advantages to both fitting the flow field to the image data and in imposing smoothness constraints between neighboring pixels. In terms of fitting, robustness is obtained to a wider variety of image motions due to the extra invariance provided by the coordinate frame. Smoothness constraints are naturally propagated along image boundaries which often correspond to motion boundaries. In addition, moving objects can be efficiently segmented in the least curvature direction. We show experimentally the benefits of the method and demonstrate robustness to fast rotational motion, such as what often occurs in human motion.
dc.description.statementofresponsibilityYan Niu, Anthony Dick and Michael Brooks
dc.identifier.citationIEEE Transactions on Image Processing, 2012; 21(4):1573-1586
dc.identifier.doi10.1109/TIP.2011.2177847
dc.identifier.issn1057-7149
dc.identifier.issn1941-0042
dc.identifier.orcidDick, A. [0000-0001-9049-7345]
dc.identifier.orcidBrooks, M. [0000-0001-9612-5884]
dc.identifier.urihttp://hdl.handle.net/2440/73419
dc.language.isoen
dc.publisherIEEE-Inst Electrical Electronics Engineers Inc
dc.relation.granthttp://purl.org/au-research/grants/arc/DP1094764
dc.relation.granthttp://purl.org/au-research/grants/arc/DP1094764
dc.rights© 2011 IEEE
dc.source.urihttps://doi.org/10.1109/tip.2011.2177847
dc.subjectDirectional derivative
dc.subjectimage structure
dc.subjectintrinsicdirection detection
dc.subjectmotion estimation
dc.subjectoptical flow
dc.titleLocally oriented optical flow computation
dc.typeJournal article
pubs.publication-statusPublished

Files

Collections

Computer Science publications