Shape Modeling with Front Propagation: A Level Set Approach
IEEE Transactions on Pattern Analysis and Machine Intelligence
A fast level set method for propagating interfaces
Journal of Computational Physics
Shape recovery algorithms using level sets in 2-D/3-D medical imagery: a state-of-the-art review
IEEE Transactions on Information Technology in Biomedicine
IEEE Transactions on Image Processing
Localized Priors for the Precise Segmentation of Individual Vertebras from CT Volume Data
MICCAI '08 Proceedings of the 11th international conference on Medical Image Computing and Computer-Assisted Intervention - Part I
| Hi-index | 0.00 |
We address the problem of precise segmentation of 3D complex structure from high-contrast images. Particularly, we focus on the representation and application of prior knowledge in the 3D level set framework. We discuss the limitations of the popular prior shape model in this type of situations, and conclude that shape model only is not complete and effective if not augmented by high-level boundary and context features. We present the principle that global priors should not compete with local image forces at the same level, but should instead guide the evolving surface to converge to the correct local primitives, thus avoiding the common problems of leakage and local minima. We propose several schemes to achieve this goal, including initial front design, speed design, and the introduction of high-level context blockers.