A case study of isosurface generation in 3D visualization

The authors explore the problem of the redundancy of the marching-cube algorithm which is widely used as an isosurface generation technique of 3-D visualization. The redundancy occurs when the pixel value from input volumetric data is equal to the isovalue chosen. A modified version of the marching-cube algorithm is also proposed. The results show that the modified algorith thus can increase the efficiency of the original algorithm in terms of memory space and execution time by removing the redundancy, e.g., a 32 /spl times/ 32 /spl times/ 32 sphere can produce 41.7% fewer triangles to represent the isosurface and 26.6% execution time can be saved.<
>