Volumetric Effect Image Processing With Machine Learning Super-Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Rendering volumetric effects in computer graphics, such as fog, smoke, or fire, is computationally expensive and often results in low quality or flickering images due to the use of low resolution sampling, which compromises temporal and spatial resolution.
Innovation Solution
A data processing apparatus that includes sampling circuitry to generate low resolution volumetric effect data, followed by super resolution circuitry using machine learning models to enhance the image resolution, and image processing circuitry to generate high-quality display images with improved temporal coherence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If volumetric rendering is performed at high resolution, then image quality is improved, but computational cost increases significantly
Solution Approach 1:
The rendering process is segmented into two stages: first generating volumetric effects at low resolution to reduce computational load, then applying a machine learning super-resolution model to enhance the output to high resolution. This segmentation allows the system to benefit from both low-resolution efficiency and high-resolution quality.
Solution Approach 2:
A machine learning super-resolution model acts as an intermediary between the low-resolution volumetric rendering output and the final high-resolution display image. This intermediary component transforms the low-resolution input into high-resolution output without requiring the original rendering to be performed at high resolution.
2Productivity
If low resolution sampling is used to reduce computational overhead, then processing speed is improved, but spatial and temporal resolution deteriorate
Solution Approach 1:
The system changes the resolution parameter dynamically: it uses low resolution during the computationally intensive volumetric rendering phase to maximize processing speed, then applies a super-resolution model to transform the low-resolution output into high-resolution final images, effectively recovering spatial and temporal resolution after the fact.
Data Source
AI summary
A data processing apparatus comprises sampling circuitry to sample computer-generated volumetric effect data for a virtual scene and generate an initial 2D volumetric effect image in dependence on a set of sampling results obtained for the computer-generated volumetric effect data; super resolution circuitry to generate a higher resolution 2D volumetric effect image in dependence on the initial 2D volumetric effect image, wherein the super resolution circuitry is configured to input the initial 2D volumetric effect image to a machine learning model trained for performing image super-resolution, the higher resolution 2D volumetric effect image having a higher image resolution than the initial 2D volumetric effect image; and image processing circuity to generate one or more display images for the virtual scene, wherein the image processing circuity is configured to generate one or more of the display images using the higher resolution 2D volumetric effect image.