Enhanced voxelization and representation of objects with sharp details in truncated distance fields

Area of Science:

• Computer Graphics
• Geometric Modeling
• Computational Geometry

Background:

• Voxelization transforms continuous objects into discrete voxel fields for rendering and analysis.
• Traditional methods can produce artifacts when objects have sharp details, failing representability conditions.
• Narrow-band distance fields offer memory efficiency and compatibility with level set methods.

Purpose of the Study:

• To present a new voxelization method for solid objects with sharp details.
• To improve the accuracy and reduce artifacts in discrete object representations.
• To enable efficient memory usage and further processing using level set techniques.

Main Methods:

• A novel approach extending distance-based voxelization.
• A priori detection and modification of sharp object details.
• Representation using truncated (narrow-band) distance fields.

Main Results:

• Successfully voxelizes solid objects with sharp details, avoiding reconstruction artifacts.
• Demonstrates effectiveness on implicit solids and Constructive Solid Geometry (CSG) operations.
• Achieves reduced memory footprints and facilitates level set computations.

Conclusions:

• The proposed method effectively handles sharp details in voxelization.
• Truncated distance fields provide an efficient representation for complex objects.
• This technique enhances the fidelity and applicability of voxel-based modeling.