Stereo Viewpoint Graphics Subsystem Sharing Geometry Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Screen-space rendering in computer graphics lacks sufficient geometric data to accurately represent three-dimensional scenes, leading to artifacts such as incorrect reflections and color bleeding due to the inability to consider off-screen surfaces, which complicates lighting and shading processes.
Innovation Solution
The implementation of a graphics processing subsystem that utilizes stereo frame buffers for left and right eye images, with a sharing decision circuit to direct the sharing of geometric data between these buffers, allowing for more accurate screen-space effect processing by leveraging the parallax between stereo viewpoints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional screen-space rendering is used with a single viewpoint, then computational efficiency is maintained, but geometric data is insufficient leading to illumination artifacts
Solution Approach 1:
The patent merges the left and right eye frame buffers by sharing geometric data between them. The sharing decision circuit determines when to share geometry data, allowing screen-space effects to access geometric information from both stereo viewpoints. This combination provides more complete geometric coverage, reducing illumination artifacts while maintaining computational efficiency through selective sharing rather than processing both buffers completely independently.
Solution Approach 2:
The patent transitions from single-viewpoint screen-space rendering to stereo viewpoint rendering, adding the dimension of parallax between left and right eye viewpoints. This dimensional change allows the system to access off-screen surfaces from the alternative viewpoint, providing missing geometric data that eliminates artifacts like incorrect reflections and color bleeding while maintaining reasonable computational overhead through the sharing mechanism.
2Loss of information
If multiple viewpoints are used to overcome screen-space limitations, then geometric data completeness improves, but processing bandwidth requirements increase
Solution Approach 1:
The patent combines information from multiple viewpoints (left and right eye frame buffers) into a unified screen-space processing approach. The sharing decision circuit enables geometric data to be merged between stereo buffers, providing complete geometric information including off-screen surfaces from either viewpoint, while avoiding the full processing overhead of independent multi-view rendering by sharing common geometric data.
Solution Approach 2:
The shared geometric data from stereo frame buffers serves multiple functions: it provides complete geometric information for accurate screen-space effects, maintains compatibility with conventional rendering pipelines, and reduces processing bandwidth requirements by avoiding redundant processing of identical geometry in both viewpoints. The sharing decision circuit enables this multi-functional use of geometric data across different rendering contexts.
3Manufacturing precision
If depth peeling or multiple cameras are used to reconstruct three-dimensional scenes, then scene detail accuracy improves, but frame buffer complexity and processing overhead increase significantly
Solution Approach 1:
The patent merges the capabilities of multiple viewpoints into the standard stereo frame buffer structure without requiring additional depth buffer complexity. By sharing geometric data between left and right eye buffers and using parallax to access off-screen surfaces, the system achieves accurate scene reconstruction equivalent to depth peeling or multiple cameras, but using the simpler existing stereo buffer architecture with minimal additional processing.
Data Source
AI summary
A stereo viewpoint graphics processing subsystem and a method of sharing geometry data between stereo images in screen-space processing. One embodiment of the stereo viewpoint graphics processing subsystem configured to render a scene includes: (1) stereo frame buffers configured to contain respective pixel-wise rendered scene data for stereo images, and (2) a sharing decision circuit operable to determine when to share geometric data between the stereo frame buffers for carrying out screen-space effect processes to render the scene in the stereo images.