Automatic method for building indoor boundary models from dense point clouds collected by laser scanners.

Area of Science:

• Computer Vision
• 3D Reconstruction
• Geometric Modeling

Background:

• Generating accurate 3D models of indoor environments is crucial for various applications.
• Existing methods often struggle with complex indoor structures and require manual intervention.

Purpose of the Study:

• To develop an automated method for generating Boundary Representation (B-rep) models of indoor spaces.
• To represent the shape, location, and relationships of primitive structural elements (walls, ceilings, floors).

Main Methods:

• Processing dense point clouds from laser scanners.
• Discretizing space for accurate 3D data segmentation.
• Generating coherent B-rep models with connected faces, edges, and vertices.

Main Results:

• Successfully generated B-rep models for indoor environments.
• Demonstrated promising results in real-world scenarios with laser scanner data.
• Evaluated the reliability of element detection and accuracy of the generated models.

Conclusions:

• The proposed method effectively automates the creation of detailed indoor 3D models.
• The approach provides accurate and coherent representations of structural elements.
• The technique shows significant potential for applications requiring precise indoor spatial data.