Motion Segmentation with Missing Data using PowerFactorization and GPCA
Description
We consider the problem of segmenting multiple rigid motions from point correspondences in multiple affine views. We cast this problem as a subspace clustering problem in which the motion of each object lives in a subspace of dimension two, three or four. Unlike previous work, we do not restrict the motion subspaces to be four-dimensional or linearly independent. Instead, our approach deals gracefully with all the spectrum of possible affine motions: from two-dimensional and partially dependent...[Show more]
dc.contributor.author | Vidal, Rene | |
---|---|---|
dc.contributor.author | Hartley, Richard | |
dc.coverage.spatial | Washington USA | |
dc.date.accessioned | 2015-12-13T22:44:59Z | |
dc.date.available | 2015-12-13T22:44:59Z | |
dc.date.created | June 27 2004 | |
dc.identifier.isbn | 0769521584 | |
dc.identifier.uri | http://hdl.handle.net/1885/79551 | |
dc.description.abstract | We consider the problem of segmenting multiple rigid motions from point correspondences in multiple affine views. We cast this problem as a subspace clustering problem in which the motion of each object lives in a subspace of dimension two, three or four. Unlike previous work, we do not restrict the motion subspaces to be four-dimensional or linearly independent. Instead, our approach deals gracefully with all the spectrum of possible affine motions: from two-dimensional and partially dependent to four-dimensional and fully independent. In addition, our method handles the case of missing data, meaning that point tracks do not have to be visible in all images. Our approach involves projecting the point trajectories of all the points into a 5-dimensional space, using the PowerFactorization method to fill in missing data. Then multiple linear subspaces representing independent motions are fitted to the points in R5 using GPCA. We test our algorithm on various real sequences with degenerate and nondegenerate motions, missing data, perspective effects, transparent motions, etc. Our algorithm achieves a misclassificatian error of less than 5% for sequences with up to 30% of missing data points. | |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE Inc) | |
dc.relation.ispartofseries | Computer Vision and Pattern Recognition Conference (CVPR 2004) | |
dc.source | Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition | |
dc.source.uri | http://cvl.umiacs.umd.edu/conferences/cvpr2004/ | |
dc.subject | Keywords: Motion segmentation; Multibody trifocal tensors; Power factorization; Subspace clustering; Algorithms; Computer vision; Data reduction; High speed cameras; Image reconstruction; Matrix algebra; Motion compensation; Motion estimation; Polynomials; Tensors; | |
dc.title | Motion Segmentation with Missing Data using PowerFactorization and GPCA | |
dc.type | Conference paper | |
local.description.notes | Imported from ARIES | |
local.description.refereed | Yes | |
dc.date.issued | 2004 | |
local.identifier.absfor | 080104 - Computer Vision | |
local.identifier.ariespublication | MigratedxPub7967 | |
local.type.status | Published Version | |
local.contributor.affiliation | Vidal, Rene, Johns Hopkins University | |
local.contributor.affiliation | Hartley, Richard, College of Engineering and Computer Science, ANU | |
local.bibliographicCitation.startpage | 310 | |
local.bibliographicCitation.lastpage | 316 | |
dc.date.updated | 2015-12-11T10:18:42Z | |
local.identifier.scopusID | 2-s2.0-5044228773 | |
Collections | ANU Research Publications |
Download
Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.
Updated: 17 November 2022/ Responsible Officer: University Librarian/ Page Contact: Library Systems & Web Coordinator