Texture mapping based on photogrammetric reconstruction of the coded markers

Area of Science:

• Computer Vision
• Computer Graphics
• 3D Modeling

Background:

• Photorealistic 3D model generation relies heavily on texture mapping.
• Existing texture mapping methods are often dependent on the features of the texture and geometric model.

Purpose of the Study:

• To introduce a novel approach for texture mapping that overcomes limitations of feature-dependent methods.
• To enhance the accuracy and robustness of texture mapping using coded markers.

Main Methods:

• Utilized coded markers as control points for texture mapping.
• Reconstructed 3D marker coordinates using photogrammetry.
• Optimized texture camera parameters via bundle adjustment.
• Calculated texture mapping pose parameters using marker registration and iterative closest point (ICP) algorithms.

Main Results:

• Successfully demonstrated the validity of the proposed algorithm through an experimental evaluation.
• The method effectively assists texture mapping by providing reliable control points.
• Achieved accurate pose parameter calculation for texture mapping.

Conclusions:

• The proposed algorithm offers a robust solution for texture mapping in 3D model generation.
• Coded markers provide a reliable alternative to feature-based methods, enhancing photorealism.
• The integration of photogrammetry, bundle adjustment, and ICP algorithms ensures accurate results.