Temporally Stable Hashed Alpha Testing for Rendering Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional alpha-tested geometry techniques in image rendering often result in undesirable visual artifacts such as flickering and disappearing geometry due to alpha variance reduction from filtering, which can lead to performance penalties and inaccurate rendering.
Innovation Solution
The implementation of hashed alpha testing, which uses a temporally stable hash function to generate consistent alpha thresholds for alpha testing, ensuring determinism, a defined output range, and uniformity, thereby reducing flickering and disappearing geometry artifacts by anchoring hash input values to 3D scene geometry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional alpha-tested geometry techniques are used to improve rendering performance, then rendering speed is improved, but visual artifacts such as flickering and disappearing geometry occur
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing hash values for discretized positions in a lookup table before rendering. During rendering, these pre-computed hash values are retrieved and used to determine alpha thresholds, eliminating the need for real-time hash computation and ensuring temporal stability across frames. This resolves the contradiction by maintaining visual stability through pre-computed consistent values while preserving rendering speed through efficient table lookup.
Solution Approach 2:
The patent uses copying by creating a discrete representation (hash table) of the continuous 3D space. Instead of computing hash values on-the-fly during rendering, the system pre-generates and stores hash values for all possible discretized positions, then copies the appropriate pre-computed value during rendering based on the current fragment position. This ensures temporal consistency while maintaining performance.
2Measurement precision
If alpha filtering is applied to reduce alpha variance, then transparency accuracy is improved, but flickering and disappearing geometry artifacts worsen
Solution Approach 1:
The patent applies local quality by using localized discretization where the hash function divides 3D space into discrete cells based on the primitive's bounding box. Each fragment's alpha threshold is determined by its specific discretized position within this local grid, allowing different regions to have different stable alpha values. This resolves the contradiction by maintaining local transparency accuracy while ensuring global temporal stability through consistent discretization.
Solution Approach 2:
The patent changes parameters by transforming the continuous alpha threshold into a discrete, quantized value based on the hash of the discretized position. Instead of using continuous alpha values that may vary slightly between frames, the system uses discrete hash-based thresholds that remain constant for the same discretized position, eliminating flickering while preserving transparency accuracy through the hash function's uniform distribution properties.
3Productivity
If simple alpha thresholding is used to maintain rendering performance, then processing speed is maintained, but rendering quality deteriorates due to flickering and disappearing geometry
Solution Approach 1:
The patent introduces an intermediary hash function that maps discretized 3D positions to alpha thresholds. This intermediary layer sits between the simple alpha thresholding approach and the desired high-quality rendering, transforming simple discrete thresholds into visually stable and accurate alpha values through the hash function's properties. This resolves the contradiction by maintaining processing speed through simple table lookup while improving rendering quality through the hash-based alpha determination.
Data Source
AI summary
A method for generating temporally stable hash values reduces visual artifacts associated with stochastic sampling of data for graphics applications. A given hash value can be generated from a scaled and discretized object-space for a geometric object within a scene. Through appropriate scaling, the hash value can be discretized and remain constant within a threshold distance from a pixel center. As the geometric object moves within the scene, a hash value associated with a given feature of the geometric object remains constant because the hash value is generated using an object-space coordinate anchored to the feature. In one embodiment, alpha testing threshold values are assigned random, but temporally stable hash output values generated using object-space coordinate positions for primitive fragments undergoing alpha testing. Alpha tested fragments are temporally stable, beneficially improving image quality.