Constraints on deformable models: recovering 3D shape and nongrid motion
Artificial Intelligence
Recursive filtering and edge closing: two primary tools for 3-D edge detection
ECCV 90 Proceedings of the first european conference on Computer vision
Active shape models—their training and application
Computer Vision and Image Understanding
A signal processing approach to fair surface design
SIGGRAPH '95 Proceedings of the 22nd annual conference on Computer graphics and interactive techniques
Globally constrained deformable models for 3D object reconstruction
Signal Processing - Special issue on deformable models and techniques for image and signal processing
IEEE Transactions on Pattern Analysis and Machine Intelligence
A level set algorithm for minimizing the Mumford-Shah functional in image processing
VLSM '01 Proceedings of the IEEE Workshop on Variational and Level Set Methods (VLSM'01)
Proceedings of the 2004 Eurographics/ACM SIGGRAPH symposium on Geometry processing
Fast streaming 3D level set segmentation on the GPU for smooth multi-phase segmentation
Transactions on computational science XIII
An efficient euclidean distance transform
IWCIA'04 Proceedings of the 10th international conference on Combinatorial Image Analysis
Adaptive Cross-sections of Anatomical Models
Computer Graphics Forum
| Hi-index | 0.00 |
In this paper, we propose a novel strategy to automatically segment volume data using a high-quality mesh segmentation of an ''example'' model as a guiding example. The example mesh is deformed until it matches the relevant volume features. The algorithm starts from a medical volume model (scalar field of densities) to be segmented, together with an already existing segmentation (polygonal mesh) of the same organ, usually from a different person. The pre-process step computes a suitable attracting scalar field in the volume model. After an approximate 3D registration between the example mesh and the volume (this is the only step requiring user intervention), the algorithm works by minimizing an energy and adapts the shape of the polygonal mesh to the volume features in order to segment the target organ. The resulting mesh adapts to the volume features in the areas which can be unambiguously segmented, while taking the shape of the example mesh in regions which lack relevant volume information. The paper discusses several examples involving human foot bones, with results that clearly outperform present segmentation schemes.