3D Graphics Indirect Illumination via Lighting-Driven Voxel Compaction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing 3D rendering algorithms fail to simultaneously achieve high-resolution images and efficient memory usage, particularly when dealing with multiple light sources, leading to increased memory consumption and decreased rendering performance.

Innovation Solution

The method introduces lighting-driven voxels (LDVs), a subset of voxels that only store reflective radiance and Morton codes for voxels contributing to indirect illumination, reducing memory usage by sorting and compacting the voxel list, and integrating this approach with voxel ray tracing and cone tracing algorithms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional voxel-based rendering algorithms are used to achieve high-resolution images, then image quality is improved, but memory consumption increases significantly

Engineering Contradiction:
Improveimage resolutionVSAvoidmemory consumption
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts and stores only the essential lighting information (reflective radiance) from voxels that contribute to indirect illumination, rather than storing complete voxel data. This selective extraction of useful information reduces memory consumption while maintaining rendering quality

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the voxel list by sorting voxels based on their contribution to indirect illumination and separating them into a compacted list of potentially lighting-driven voxels (pLDVs). This segmentation allows the system to process only relevant voxels, reducing memory requirements for high-resolution rendering

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If traditional rendering algorithms process all voxels from multiple light sources, then illumination accuracy is improved, but processing time increases

Engineering Contradiction:
Improveillumination accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary sorting and compaction of the voxel list before rendering, organizing voxels by their lighting contribution in advance. This preliminary action enables faster processing during actual rendering by eliminating the need to process irrelevant voxels, thus reducing processing time while maintaining illumination accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent processes only the subset of voxels that are potentially lighting-driven (pLDVs) rather than all voxels in the scene. This partial processing approach maintains illumination accuracy for visible surfaces while significantly reducing the computational burden and processing time

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS10275937B2Indirect illumination method and 3D graphics processing device
Publication Date: 2019.04.30 MEDIATEK INC
  • US10275937B2 patent drawing
  • US10275937B2 patent drawing
  • US10275937B2 patent drawing

AI summary

The present invention provides a method of indirect illumination, for a 3D graphics processing device, including obtaining a scene and perform a voxelization to the scene; performing a lighting computation to the voxelized scene from a plurality of light sources, and store a potential lighting driven voxel (pLDV) list according to the lighting computation; sorting the pLDV list to generate a sorted pLDV list; and performing a compaction process to the sorted pLDV list; wherein each voxel in the pLDV list stores a reflective radiance and a Morton code corresponding to each voxel.