Discrete and Continuous Optimization for Motion Estimation

Simple item page
dc.contributor.advisorCamillo J. Taylor
dc.contributor.authorKennedy, Ryan
dc.date2023-05-17T11:57:05.000
dc.date.accessioned2023-05-22T16:21:26Z
dc.date.available2001-01-01T00:00:00Z
dc.date.copyright2015-07-20T20:15:00-07:00
dc.date.issued2015-01-01
dc.date.submitted2015-07-20T07:06:05-07:00
dc.description.abstractThe study of motion estimation reaches back decades and has become one of the central topics of research in computer vision. Even so, there are situations where current approaches fail, such as when there are extreme lighting variations, significant occlusions, or very large motions. In this thesis, we propose several approaches to address these issues. First, we propose a novel continuous optimization framework for estimating optical flow based on a decomposition of the image domain into triangular facets. We show how this allows for occlusions to be easily and naturally handled within our optimization framework without any post-processing. We also show that a triangular decomposition enables us to use a direct Cholesky decomposition to solve the resulting linear systems by reducing its memory requirements. Second, we introduce a simple method for incorporating additional temporal information into optical flow using "inertial estimates" of the flow, which leads to a significant reduction in error. We evaluate our methods on several datasets and achieve state-of-the-art results on MPI-Sintel. Finally, we introduce a discrete optimization framework for optical flow computation. Discrete approaches have generally been avoided in optical flow because of the relatively large label space that makes them computationally expensive. In our approach, we use recent advances in image segmentation to build a tree-structured graphical model that conforms to the image content. We show how the optimal solution to these discrete optical flow problems can be computed efficiently by making use of optimization methods from the object recognition literature, even for large images with hundreds of thousands of labels.
dc.description.degreeDoctor of Philosophy (PhD)
dc.format.extent156 p.
dc.format.mimetypeapplication/pdf
dc.identifier.urihttps://repository.upenn.edu/handle/20.500.14332/27849
dc.languageen
dc.legacy.articleid2885
dc.legacy.fulltexturlhttps://repository.upenn.edu/cgi/viewcontent.cgi?article=2885&context=edissertations&unstamped=1
dc.provenanceReceived from ProQuest
dc.rightsRyan Kennedy
dc.source.issue1075
dc.source.journalPublicly Accessible Penn Dissertations
dc.source.statuspublished
dc.subject.otherComputer Vision
dc.subject.otherMotion
dc.subject.otherOptical Flow
dc.subject.otherComputer Sciences
dc.titleDiscrete and Continuous Optimization for Motion Estimation
dc.typeDissertation/Thesis
digcom.contributor.authorisAuthorOfPublication|email:kenryd@gmail.com|institution:University of Pennsylvania|Kennedy, Ryan
digcom.date.embargo2001-01-01T00:00:00-08:00
digcom.identifieredissertations/1075
digcom.identifier.contextkey7345429
digcom.identifier.submissionpathedissertations/1075
digcom.typedissertation
dspace.entity.typePublication
relation.isAuthorOfPublication32e9362a-cc9c-4fb8-96dd-be3036df75e7
relation.isAuthorOfPublication.latestForDiscovery32e9362a-cc9c-4fb8-96dd-be3036df75e7
upenn.graduate.groupComputer and Information Science
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Kennedy_upenngdas_0175C_11607.pdf
Size:
89.25 MB
Format:
Adobe Portable Document Format
Download
Collection
Dissertations and Theses