Adaptive Bias Shadow Mapping for Real-Time Graphics Rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating shadows in real-time graphics rendering are computationally intensive and suffer from aliasing artifacts due to finite shadow map resolution, requiring view-dependent regeneration and leading to erroneous shadowing and shifting issues.
Innovation Solution
A method involving computing a depth-map with adaptive bias based on surface orientation and light direction to accurately label pixels as lit or shadowed, using a graphics processing unit to project points into light space, calculate distances, and apply an adaptive bias for accurate shadow determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If shadow volumes method is used to generate high quality shadows, then shadow quality is improved, but computational complexity increases and it is limited to simple scenes
Solution Approach 1:
The patent uses shadow mapping which creates a 2D depth map (copy) of the 3D scene from the light source perspective. This copy is then used to determine shadowing in the final image, replacing the computationally intensive shadow volumes method while maintaining acceptable shadow quality.
Solution Approach 2:
The patent changes the approach from calculating actual shadow volumes in 3D space to using a 2D depth map with depth values. This parameter change from 3D geometric computation to 2D texture sampling significantly reduces computational complexity while producing comparable shadow results.
2Productivity
If shadow map technique is used for real-time shadow generation, then computational efficiency is improved, but aliasing artifacts appear due to finite resolution
Solution Approach 1:
The patent applies percentage-closer filtering which changes the sampling parameters by using multiple shadow map texels and computing a weighted average based on their distances. This parameter change from single-texel sampling to multi-texel filtering reduces aliasing artifacts while maintaining real-time performance.
3Manufacturing precision
If view-dependent shadow map regeneration is implemented to reduce aliasing artifacts, then shadow quality is improved, but computational time increases and user movement causes shadow shifting
Solution Approach 1:
The patent pre-calculates and stores depth values in a shadow map texture before rendering the final scene. This preliminary action allows the shadow information to be readily available during real-time rendering without requiring computation at the moment of view change, thus avoiding delays and shadow shifting artifacts.
Solution Approach 2:
The shadow map is designed to be view-independent and can be reused across multiple view angles without regeneration. This universal shadow map serves multiple viewing perspectives simultaneously, eliminating the need for view-dependent regeneration and its associated performance penalties.
4Manufacturing precision
If user-defined bias value is applied to offset distance comparisons, then erroneous shadowing is removed, but all shadows shift undesirably
Solution Approach 1:
The patent implements per-pixel bias calculation where each pixel receives a bias value tailored to its specific geometric context (surface orientation relative to light direction). This local quality approach ensures that bias is applied only where geometric errors occur rather than uniformly across all shadows, preventing unwanted shadow shifting while correcting local artifacts.
Data Source
AI summary
As to generate shadows in an image, the method comprises the steps of:Computing a depth-map that comprises an array of pixels, wherein pixels in the depth-map are associated to a single value corresponding to depth value that indicates a depth from a light source to a portion of nearest occluding object visible through the pixel,projecting a point visible through a pixel of said image into a light space, the result of said projection being a pixel of said depth-map,calculating a distance between the said visible point and the light source,fetching the depth value associated to said pixel of depth-map,computing, for said pixel of said image, an adaptive bias as a function of a predetermined base bias and a relationship between the normal of a surface on which the said visible point is located and incident light direction at said visible point,comparing for said pixel in the image, the distance between said visible point and the light source with the sum of the corresponding depth map value and said adaptive bias,labelling said point visible through said pixel as lit or shadowed according to said comparison.