Dynamic Rendering Model Selection for Frame Rate Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing rendering systems struggle to maintain target frame rates and realism in complex virtual environments, often resorting to noticeable reductions in resolution and frame rate, which degrade the immersiveness of the content.
Innovation Solution
A machine learning model is trained to select the most optimal rendering model in real-time based on the static and dynamic properties of the processor, adapting to changing load conditions to maintain image quality and efficiency without altering resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the resolution is reduced to bring the frame rate back up to the target rate, then the frame rate is improved, but the image quality and realism deteriorate
Solution Approach 1:
The system dynamically switches between different rendering models (first rendering model and second rendering model) based on real-time processor load conditions. When the processor load is high, it switches to a more efficient rendering model to maintain frame rate, and when load is low, it uses a higher quality rendering model, making the system adaptable to changing conditions rather than using a fixed resolution reduction approach
Solution Approach 2:
The invention changes the rendering model parameter based on processor load rather than changing the resolution parameter. This allows the system to maintain consistent resolution while adjusting computational complexity through model selection, thereby improving frame rate without sacrificing image quality
2Device complexity
If a fixed rendering model is used, then the device complexity is reduced, but the adaptability to changing load conditions deteriorates
Solution Approach 1:
The rendering system incorporates dynamic model switching capability that adapts to changing processor load conditions. The system monitors load metrics and automatically selects between rendering models, providing adaptability while keeping the switching logic integrated within the existing rendering pipeline rather than requiring separate complex control systems
Solution Approach 2:
The rendering system performs self-adjustment by automatically selecting appropriate rendering models based on monitored processor load conditions. The system services its own adaptation needs through integrated load monitoring and model selection logic, eliminating the need for external complex control mechanisms
3Productivity
If the resolution is reduced temporarily, then the frame rate is improved, but the user experience and immersiveness deteriorate
Solution Approach 1:
Instead of changing the resolution parameter, the system changes the rendering model parameter to adjust computational complexity. This maintains consistent image quality and resolution that users expect, while improving frame rate through more efficient rendering algorithms selected based on processor load
Solution Approach 2:
The system dynamically adjusts rendering model selection to maintain consistent visual quality across varying load conditions. By switching between rendering models rather than resolution levels, the system preserves the visual experience and immersiveness while adapting frame rate performance to available processing resources
Data Source
AI summary
The image processing method comprises: rendering, by a rendering processor, a first image frame of content using a first rendering model; and receiving input data relating to a plurality of properties of the rendering processor. The rendering processor comprising one or more static properties and one or more dynamic properties indicative of a current load on the rendering processor. The method further comprises: inputting the input data to a machine learning model trained to select a rendering model, amongst a plurality of rendering models, for use in rendering an image frame, in dependence on the properties of a processor for rendering the image frame; selecting, by the trained machine learning model, a second rendering model in dependence on the input data; and rendering, by the rendering processor, at least part of a second image frame of the content using the second rendering model in place of the first rendering model.