Abstract
Groupwise non-rigid registration is an important technique in medical image analysis. Recent studies show that its accuracy can be greatly improved by explicitly providing good initialisation. This is achieved by seeking a sparse correspondence using a parts+geometry model. In this paper we show that a single parts+geometry model is unlikely to establish consistent sparse correspondence for complex objects, and that better initialisation can be achieved using a set of models. We describe how to combine the strengths of multiple models, and demonstrate that the method gives state-of-the-art performance on three datasets, with the most significant improvement on the most challenging.
Original language | English |
---|---|
Title of host publication | Medical Image Computing and Computer-Assisted Intervention, MICCAI2012 - 15th International Conference, Proceedings |
Editors | Nicholas Ayache, Herve Delingette, Polina Golland, Kensaku Mori |
Publisher | Springer Verlag |
Pages | 156-163 |
Number of pages | 8 |
ISBN (Print) | 9783642334535 |
DOIs | |
Publication status | Published - 2012 |
Event | 15th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2012 - Nice, France Duration: 2012 Oct 1 → 2012 Oct 5 |
Publication series
Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Volume | 7512 LNCS |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | 15th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2012 |
---|---|
Country/Territory | France |
City | Nice |
Period | 12/10/1 → 12/10/5 |
Bibliographical note
Publisher Copyright:© Springer-Verlag Berlin Heidelberg 2012.
ASJC Scopus subject areas
- Theoretical Computer Science
- General Computer Science