In this paper, we propose a new multiresolution-based shape representation for 3 0 mesh morphing. Our approach does not use combination operations that caused some serious problems in the previous approaches for mesh morphing. Therefore, we can calculate a hierarchical interpolation mesh robustly using two types of subdivision jitting schemes. Our new representation has a hierarchical semiregular mesh structure based on subdivision connectivi9. This leads to various advantages including eficient data storage, and easy acquisition of an interpolation mesh with arbitrary subdivision level. We also demonstrate several new features for 3 0 morphing using multiresolution interpolation meshes.
The authors present a method for extracting polygon data of endocranial surfaces from CT images of human crania. Based on the fact that the endocast is the largest empty space in the crania, we automate a procedure for endocast extraction by integrating several image processing techniques. Given CT images of human crania, the proposed method extracts endocranial surfaces by the following three steps. The first step is binarization in order to fill void structures, such as diploic space and cracks in the skull. We use a void detection method based on mathematical morphology. The second step is watershed-based segmentation of the endocranial part from the binary image of the CT image. Here, we introduce an automatic initial seed assignment method for the endocranial region using the distance field of the binary image. The final step is partial polygonization of the CT images using the segmentation results as mask images. The resulting polygons represent only the endocranial part, and the closed manifold surfaces are computed even though the endocast is not isolated in the cranium. Since only the isovalue threshold and the size of void structures are required, the procedure is not dependent on the experience of the user. The present paper also demonstrates that the proposed method can extract polygon data of endocasts from CT images of various crania.
a) Input CT image of an object (b) Distance field on the inside of the object (c) Distance field on the outside of the object Figure 1: The result of the OoCDT algorithm for an "Cylinder-Head" model (size: 850 × 800 × 1059 ≈ 700M cells). Note that the data size of the distance field is about 3GB (4 bytes/cell), which is too large to allocate to the RAM. Our algorithm can compute such large distance fields on common 32-bit computers..
AbstractThis paper presents a method for computing distance fields from large volumetric models. Conventional methods have strict limits in terms of the amount of memory space available, as all volumetric models must be allocated to the random access memory (RAM) to compute distance fields. We resolve this problem through an outof-core strategy. Our algorithm starts by decomposing volumetric models into small regions known as clusters, and distance fields are then computed by Local Distance Transform (LDT) and InterCluster Propagation (ICP). LDT computes the distance transform for each cluster, and since it is independent, other clusters can also be saved to the storage medium. ICP propagates the distance at the boundary of the cluster to neighboring clusters to remove inconsistency in distance fields. In addition, we propose an efficient ordering algorithm based on the propagated distance to reduce LDT and ICP. This paper also demonstrates the results of distance transform from volumetric models with over a billion cells.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.