Publications
Text queries can be conducted by Author, Title, or Keyword.
Distance field manipulation of surface models
Source: IEEE Computer Graphics and Applications 1992 Feb;12(1):65-71.
Author: Payne BA, Toga AW.
Abstract:
Scientific visualization frequently gives rise to complex models of surfaces. In biomedical imaging, organs such as the brain provide elaborate structures to image and analyze. Surfaces might come from interactive techniques, such as outlining, or automatic ones like isosurface creation. Surface models of simpler objects (such as cubes and spheres) are also common, and you can use a number of techniques to image such models. But manipulating surfaces using either direct or implicit methods presents a number of challenges. To manipulate surfaces more effectively, we developed a method that uses distance fields-the scalar fields derived from triangle-based surface models. We face four specific challenges when trying to manipulate the surfaces in a visualization model: (1) Averaging/interpolation: What is an average of two or more surfaces? Given models of different objects, what is the typical or representative surface? Finding a representative surface allows interpolation, continuously deforming one surface into another. (2) Offsets: What is the surface a fixed distance above or below a given distance? What is the surface a fixed proportion between two given surfaces? The cortex of the brain, for example, is organized into layers that might by accessible by offset surfaces. (3) Blending: How can we combine multiple surfaces in one view with continuous joins between adjacent objects? This combining is similar to what we do with blobby objects in implicit surface modeling or blends and fillets in CAD/CAM. (4) Blurring: How can we represent a surface with less resolution or detail while preserving large-scale features? For example, can we remove high frequency errors while preserving lower frequency information?
Source: IEEE Computer Graphics and Applications 1992 Feb;12(1):65-71.
Author: Payne BA, Toga AW.
Abstract:
Scientific visualization frequently gives rise to complex models of surfaces. In biomedical imaging, organs such as the brain provide elaborate structures to image and analyze. Surfaces might come from interactive techniques, such as outlining, or automatic ones like isosurface creation. Surface models of simpler objects (such as cubes and spheres) are also common, and you can use a number of techniques to image such models. But manipulating surfaces using either direct or implicit methods presents a number of challenges. To manipulate surfaces more effectively, we developed a method that uses distance fields-the scalar fields derived from triangle-based surface models. We face four specific challenges when trying to manipulate the surfaces in a visualization model: (1) Averaging/interpolation: What is an average of two or more surfaces? Given models of different objects, what is the typical or representative surface? Finding a representative surface allows interpolation, continuously deforming one surface into another. (2) Offsets: What is the surface a fixed distance above or below a given distance? What is the surface a fixed proportion between two given surfaces? The cortex of the brain, for example, is organized into layers that might by accessible by offset surfaces. (3) Blending: How can we combine multiple surfaces in one view with continuous joins between adjacent objects? This combining is similar to what we do with blobby objects in implicit surface modeling or blends and fillets in CAD/CAM. (4) Blurring: How can we represent a surface with less resolution or detail while preserving large-scale features? For example, can we remove high frequency errors while preserving lower frequency information?