Fast Classification in Space-Leaping for Virtual Endoscopy

초록

Several volume rendering methods such as space-leaping and shear-warp rendering generate spatial data structure in preprocessing time for speedup. Although they are useful for improving rendering speed, it may take a lot of time to regenerate them for some reason, for example, change of opacity transfer function. In this paper we propose a high-speed classification method for space-leaping. We should perform distance transformation for all voxels in the previous method. Instead of regenerating entire distance-map, our method modifies the values of some part of entire map. That is, when the opacity transfer function is altered, pre-defined values are assigned to voxels of which transparency is changed. Our rendering algorithm determines the next sampling position by interpreting the values of those tagged voxels in special manner. Therefore we can adjust transparency of each voxel in moderate speed, and render volume data sets without loss of image quality.