Virtual Scene Lighting Rendering via Contribution-Weighted Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional lighting rendering in virtual scenes requires uniform sampling of lighting paths, leading to poor rendering quality due to the need for a large number of sampled paths, especially when the number of paths is small.
Innovation Solution
A method and apparatus for lighting rendering that selects candidate lighting paths based on their respective probabilities correlated with lighting contribution values, incorporating first and second lighting paths from the pixel point and its neighborhood, to determine a comprehensive lighting contribution value for improved rendering quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If uniform sampling is used to acquire lighting paths, then the sampling process is simple, but a large number of paths must be sampled to achieve convergence, resulting in poor rendering quality when the number of sampled paths is small
Solution Approach 1:
The patent changes the sampling parameter from uniform probability distribution to non-uniform probability distribution weighted by lighting contribution values. This allows the sampling process to focus on more important lighting paths while maintaining computational feasibility, thereby improving rendering quality without requiring an excessively large number of samples.
Solution Approach 2:
The patent applies different sampling strategies to different regions of the light source. By calculating lighting contribution values for different areas and sampling more densely in regions with higher contribution, the method achieves better local rendering quality in critical areas while maintaining overall efficiency.
2Manufacturing precision
If a large number of lighting paths are sampled to improve rendering quality, then rendering quality improves, but the computational time and resources increase significantly
Solution Approach 1:
By changing from uniform sampling to contribution-weighted sampling, the patent achieves better rendering quality with fewer samples. The probability distribution is adjusted based on lighting contribution values, allowing the system to converge faster and reduce computational time while maintaining or improving quality.
Solution Approach 2:
The patent samples more lighting paths than the minimum required for basic convergence, but focuses this sampling effort on paths with higher contribution values. This partial emphasis on critical paths achieves superior quality without the proportional time cost of uniformly sampling all paths.
3Ease of operation
If light rays are emitted in various directions of pixel points at the same probability, then the sampling is uniform and simple, but poor rendering quality is caused when the number of sampled lighting paths is small
Solution Approach 1:
The patent modifies the probability parameter from uniform distribution to non-uniform distribution based on lighting contribution. This allows the sampling to remain systematic and manageable while achieving superior rendering quality by directing more samples toward important lighting directions.
Solution Approach 2:
Different sampling probabilities are assigned to different directions based on their lighting contribution to the pixel. Directions that contribute more to the final image receive higher sampling probability, creating local quality enhancement in critical viewing directions while maintaining overall operational simplicity.
Data Source
AI summary
Lighting rendering techniques are described herein. The techniques may include: determining, for each pixel point in a virtual scene, a plurality of first lighting paths acquired through lighting path sampling on the pixel point, and selecting a first candidate lighting path from the plurality of first lighting paths; determining a neighboring pixel point, and determining a plurality of second lighting paths acquired through lighting path sampling on the neighboring pixel point selecting a target lighting path from the first candidate lighting path and a second candidate lighting path; determining a related lighting path of the pixel point, determining, with reference to the lighting contribution value of the targeted lighting path, a lighting contribution of the related lighting path to the pixel point, and acquiring a comprehensive lighting contribution value of the pixel point; and performing the lighting rendering on each pixel point in the virtual scene using the comprehensive lighting contribution value of each pixel point in the virtual scene.