Virtual Scene Illumination Rendering with Importance Path Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional lighting rendering methods in virtual scenes require 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 that select candidate lighting paths based on their positively correlated lighting contribution values, determining related lighting paths from both the pixel point and its neighborhood, and using these paths to compute comprehensive lighting contribution values 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 the rendering quality is poor due to the need for a large number of sampled paths
Solution Approach 1:
The patent changes the sampling parameter from uniform random sampling to importance sampling based on lighting contribution values. By calculating the lighting contribution value for each sampled path and using it as a weight for re-sampling, the method concentrates sampling effort on more important paths, thereby improving rendering quality without requiring an excessive number of samples.
Solution Approach 2:
The patent implements a feedback mechanism where the lighting contribution values obtained from initial sampling are used to guide subsequent sampling processes. The contribution values are fed back to determine the probability distribution for re-sampling, creating an iterative optimization that continuously improves rendering quality.
2Manufacturing precision
If a large number of lighting paths are sampled to improve rendering quality, then the rendering quality improves, but the computational efficiency decreases
Solution Approach 1:
The patent changes the sampling strategy from uniform sampling requiring many paths to importance sampling that achieves better quality with fewer paths. By using lighting contribution values as weights, the method efficiently allocates sampling resources to the most impactful paths, improving both rendering quality and computational efficiency simultaneously.
Solution Approach 2:
The patent applies partial action by selecting only the top-k lighting paths with the highest contribution values for final rendering, rather than using all sampled paths. This selective approach maintains rendering quality while significantly reducing the computational burden of processing every sampled path.
3Ease of manufacture
If uniform sampling is used, then the implementation is straightforward, but the convergence requires a large number of sampled paths
Solution Approach 1:
The patent modifies the sampling parameter from uniform distribution to a non-uniform distribution based on lighting contribution values. This change allows the system to converge faster with fewer sampled paths, as the importance sampling naturally directs attention to the most significant lighting contributions without requiring excessive sample quantities.
Data Source
Figure 1~2
Figure 3~5(c)
Figure 6~7(c2)
AI summary
Disclosed is an image rendering method. The method is performed by a computer device and includes: determining, for a 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, such that a probability of the first candidate lighting path being selected is positively correlated with a corresponding lighting contribution value of the first candidate lighting path (202); determining a neighborhood pixel point of the pixel point, determining a plurality of second lighting paths acquired through lighting path sampling on the neighborhood pixel point, and selecting a second candidate lighting path from the plurality of second lighting paths, such that a probability of the second candidate lighting path being selected is positively correlated with a corresponding lighting contribution value of the second lighting path (204); selecting a target lighting path from the first candidate lighting path and the second candidate lighting path (206); determining a related lighting path of the pixel point, and determining, according to the lighting contribution value of the target lighting path, a lighting contribution that the related lighting path provides to the pixel point, to acquire a lighting contribution value of the pixel point; the related lighting path comprising the first lighting path and the second lighting path (208); and performing image rendering on the pixel point in the virtual scene by using the lighting contribution value of the pixel point in the virtual scene (210).