MSAA Buffer Upscaling With Pretrained Blending Weights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing graphics processing technologies face challenges in efficiently upscaling images to high resolution on low-power devices, requiring either proprietary hardware or high computational resources, while simple algorithms fail to deliver high-quality results.
Innovation Solution
Utilizing multi-sample anti-aliasing (MSAA) intermediate buffers to store micro-geometry information, combined with pre-trained blending weight vectors, for efficient upscaling to higher resolutions using data-driven techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If proprietary hardware or high computational resources are used for upscaling, then image quality is improved, but device complexity and power consumption increase
Solution Approach 1:
The patent uses MSAA intermediate buffers as an intermediary resource that already contains micro-geometry information from the rendering pipeline. Instead of directly upscaling from low-resolution images, the system leverages these intermediate buffers that capture edge and geometry data, enabling quality upscaling without requiring proprietary hardware. The intermediate buffer acts as a mediator between the rendering process and the upscaling operation.
Solution Approach 2:
The patent performs preliminary actions by capturing micro-geometry information during the rendering process itself through MSAA. This geometry information is stored in intermediate buffers before the upscaling operation occurs. By preparing this data in advance during rendering, the system eliminates the need for complex proprietary hardware during the actual upscaling phase, as the necessary geometric data is already available.
2Measurement precision
If proprietary hardware is used for upscaling, then image quality is improved, but power consumption increases
Solution Approach 1:
The patent implements self-service by having the rendering pipeline generate and store the necessary micro-geometry information in MSAA intermediate buffers during normal operation. The upscaling process then serves itself by utilizing this pre-captured geometry data without requiring external proprietary hardware assistance. This self-sufficient approach eliminates the power consumption associated with specialized upscaling hardware while maintaining high image quality.
3Productivity
If simple upscaling algorithms are used, then computational resources are reduced, but image quality deteriorates
Solution Approach 1:
The patent substitutes traditional mechanical upscaling algorithms with a data-driven approach that leverages MSAA intermediate buffers containing micro-geometry information. Instead of using complex iterative algorithms that consume significant computational resources, the system replaces the mechanical upscaling process with a more efficient method that utilizes pre-captured geometry data to directly generate high-quality upscaled images with simpler computations.
4Measurement precision
If high computational resources are allocated for upscaling, then image quality is improved, but runtime computation complexity increases
Solution Approach 1:
The patent performs preliminary action by capturing and storing micro-geometry information in MSAA intermediate buffers during the rendering phase. This pre-prepared geometry data eliminates the need for complex runtime computations during upscaling, as the essential geometric information is already available from the preliminary rendering process. The upscaling operation becomes simpler because it works with pre-processed geometry data rather than generating it from scratch.
Data Source
AI summary
Processing circuitry (e.g., GPU) obtains a first MSAA intermediate buffer from a rendering of a first image, the first MSAA intermediate buffer includes a plurality of subsamples for each pixel of the first image. The processing circuitry determines, for a first pixel of the first image, a first surrounding subsample pattern based on first subsamples in a surrounding region of the first pixel. Further, the processing circuitry obtains, based on the first surrounding subsample pattern, a first blending weight vector for a first location among a plurality of locations that define output pixel locations within the first pixel for upscaling the first pixel to a plurality of output pixels in an upscaled image of higher resolution than the first image. The first blending weight vector includes first blending weights respectively for first candidate subsamples associated with the first pixel. The processing circuitry generates the upscaled image accordingly.