Real-Time Rendering Methods With Adaptive Levels of Detail for Fast Rendering of Parametric Objects on Modern GPUs.

Area of Science:

• Computer Graphics
• Geometric Modeling
• GPU Computing

Background:

• Parametric functions offer efficient and compact 3D geometry representation.
• Visualizing parametric objects at arbitrary Levels of Detail (LOD) is memory-efficient but challenging for GPU rendering pipelines.

Purpose of the Study:

• To propose a general method for efficient rendering of parametrically defined 3D objects on modern hardware.
• To enable adaptive analysis, allocation, and evaluation of parametric function samples for high-quality renderings.

Main Methods:

• An adaptive LOD stage analyzes and allocates samples for efficient rendering.
• The method supports both hardware tessellation and point-based software rasterization.
• No preprocessing or caching is required, allowing real-time LOD adjustments.

Main Results:

• Achieved real-time frame rates (hundreds of FPS) with adjustable geometric precision.
• Demonstrated superior performance over spherical harmonics glyph rendering and classical LOD methods.
• Showcased flexibility across various complex parametric shapes.

Conclusions:

• The proposed method provides efficient, high-quality rendering of parametric 3D objects on modern GPUs.
• It supports arbitrary LOD and real-time performance, suitable for animated scenes.
• The approach is general, requiring no preprocessing and offering significant advantages over existing techniques.