A spectral technique for correspondence problems using pairwise constraints
Top Cited Papers
- 1 January 2005
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2 (15505499) , 1482
- https://doi.org/10.1109/iccv.2005.20
Abstract
We present an efficient spectral method for finding consistent correspondences between two sets of features. We build the adjacency matrix M of a graph whose nodes represent the potential correspondences and the weights on the links represent pairwise agreements between potential correspondences. Correct assignments are likely to establish links among each other and thus form a strongly connected cluster. Incorrect correspondences establish links with the other correspondences only accidentally, so they are unlikely to belong to strongly connected clusters. We recover the correct assignments based on how strongly they belong to the main cluster of M, by using the principal eigenvector of M and imposing the mapping constraints required by the overall correspondence mapping (one-to-one or one-to-many). The experimental evaluation shows that our method is robust to outliers, accurate in terms of matching rate, while being much faster than existing methodsKeywords
This publication has 10 references indexed in Scilit:
- Unsupervised Learning of Object Features from Video SequencesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Segmentation of multiple salient closed contours from real imagesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- A global solution to sparse correspondence problemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Alignment using Spectral ClustersPublished by British Machine Vision Association and Society for Pattern Recognition ,2002
- Object recognition from local scale-invariant featuresPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1999
- Quantitative Measures of Change Based on Feature Organization: Eigenvalues and EigenvectorsComputer Vision and Image Understanding, 1998
- ARPACK Users' GuidePublished by Society for Industrial & Applied Mathematics (SIAM) ,1998
- Feature-based correspondence: an eigenvector approachImage and Vision Computing, 1992
- A method for registration of 3-D shapesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1992
- Spline Models for Observational DataPublished by Society for Industrial & Applied Mathematics (SIAM) ,1990