Surface reconstruction from discrete indicator functions.

Area of Science:

• Computer graphics
• Geometric modeling
• Scientific visualization

Background:

• Surface reconstruction from discrete data is crucial in many scientific fields.
• Existing methods like Marching Cubes often require smooth input functions, posing challenges for discrete indicator functions.
• Linear or higher-order interpolation methods typically need smooth input, making discontinuous indicator functions difficult to process.

Purpose of the Study:

• To introduce a novel vertex interpolation method for Marching Cubes-like algorithms.
• To achieve second-order accuracy in surface reconstruction from discrete indicator functions in 3D.
• To develop a method that maintains locality and avoids the high computational cost of rigorous 3D extensions.

Main Methods:

• A new analytical procedure for calculating vertex positions in surface reconstruction.
• Development of a second-order-accurate interpolation technique for discontinuous indicator functions.
• Error analysis and comparison with existing linear interpolation and Manson et al.'s 2D analytical formulations extended to 3D.

Main Results:

• The proposed method yields second-order-accurate reconstructed surfaces in the general 3D case.
• The method preserves the locality of the Marching Cubes algorithm.
• Demonstrated superior accuracy compared to linear interpolation and Manson et al.'s approach in 3D.

Conclusions:

• The novel vertex interpolation method offers an efficient and accurate solution for reconstructing surfaces from discrete indicator functions.
• This technique enhances the quality of surface reconstruction in 3D computer graphics and scientific visualization.
• The method provides a valuable alternative to existing techniques, particularly when dealing with discontinuous data.