TY - GEN
T1 - Detecting mutually-salient landmark pairs with MRF regularization
AU - Ou, Y.
AU - Besbes, A.
AU - Bilellod, M.
AU - Mansourd, M.
AU - Davatzikos, C.
AU - Paragios, N.
PY - 2010
Y1 - 2010
N2 - In this paper, we present a framework for extracting mutually-salient landmark pairs for registration. Traditional methods detect landmarks one-by-one and separately in two images. Therefore, the detected landmarks might inherit low discriminability and are not necessarily good for matching. In contrast, our method detects landmarks pair-by-pair across images, and those pairs are required to be mutually-salient, i.e., uniquely corresponding to each other. The second merit of our framework is that, instead of finding individually optimal correspondence, which is a local approach and could cause self-intersection of the resultant deformation, our framework adopts a Markov-random-field (MRF)-based spatial arrangement to select the globally optimal landmark pairs. In this way, the geometric consistency of the correspondences is maintained and the resultant deformations are relatively smooth and topology-preserving. Promising experimental validation through a radiologist's evaluation of the established correspondences is presented.
AB - In this paper, we present a framework for extracting mutually-salient landmark pairs for registration. Traditional methods detect landmarks one-by-one and separately in two images. Therefore, the detected landmarks might inherit low discriminability and are not necessarily good for matching. In contrast, our method detects landmarks pair-by-pair across images, and those pairs are required to be mutually-salient, i.e., uniquely corresponding to each other. The second merit of our framework is that, instead of finding individually optimal correspondence, which is a local approach and could cause self-intersection of the resultant deformation, our framework adopts a Markov-random-field (MRF)-based spatial arrangement to select the globally optimal landmark pairs. In this way, the geometric consistency of the correspondences is maintained and the resultant deformations are relatively smooth and topology-preserving. Promising experimental validation through a radiologist's evaluation of the established correspondences is presented.
KW - Image registration
KW - Landmark matching
KW - MRF
KW - Mutual-saliency
UR - http://www.scopus.com/inward/record.url?scp=77955188044&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77955188044&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2010.5490324
DO - 10.1109/ISBI.2010.5490324
M3 - Conference contribution
AN - SCOPUS:77955188044
SN - 9781424441266
T3 - 2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings
SP - 400
EP - 403
BT - 2010 7th IEEE International Symposium on Biomedical Imaging
PB - IEEE Computer Society
T2 - 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010
Y2 - 14 April 2010 through 17 April 2010
ER -