Title | ||
---|---|---|
Polyrigid and Polyaffine Transformations: A New Class of Diffeomorphisms for Locally Rigid or Affine Registration |
Abstract | ||
---|---|---|
We introduce a novel family of geometrical transformations for the non-rigid registration of medical images, called polyrigid and polyaffine. These transformations have nice mathematical properties, including differentiability and invertibility (i.e diffeomorphism). They are parameterized by a small number of intuitive parameters which allow to recover locally rigid and locally affine deformations often encountered in medical imagery. These transformations are a nice alternative to classical B-Spline transformations (which do not guaranty invertibility). Preliminary experiments on synthetic and real images already illustrate the potentialities and good properties of this approach. |
Year | DOI | Venue |
---|---|---|
2003 | 10.1007/978-3-540-39903-2_101 | Lecture Notes in Computer Science |
Field | DocType | Volume |
Affine transformation,Computer vision,Parameterized complexity,Computer science,Differentiable function,Artificial intelligence,Real image,Mathematical properties,Diffeomorphism | Conference | 2879 |
ISSN | Citations | PageRank |
0302-9743 | 13 | 1.12 |
References | Authors | |
12 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Vincent Arsigny | 1 | 733 | 50.69 |
Xavier Pennec | 2 | 5021 | 357.08 |
Nicholas Ayache | 3 | 10804 | 1654.36 |