VR Scene Rendering via Pre-rendered Image Caching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Virtual reality devices experience dizziness due to long time delays caused by low GPU rendering refresh rates, limiting display refresh rates and frame rates.
Innovation Solution
Rendering images in an idle state and using a calling method during display to improve the display refresh rate, reduce delays, and enhance image authenticity, while reducing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-time GPU rendering is used to maintain image authenticity, then image quality is improved, but display refresh rate is limited due to low rendering speed
Solution Approach 1:
The patent applies preliminary action by pre-rendering images during idle states before they are needed for display. The system determines when the GPU is idle and proactively renders images for future display frames, so that when display is needed, pre-rendered images are already available. This resolves the contradiction by preparing images in advance during idle periods, enabling high display refresh rates without compromising image authenticity during active display operations.
Solution Approach 2:
The patent implements dynamics by dynamically switching between real-time rendering and pre-rendered image display based on system state. The system monitors GPU availability and dynamically adjusts the rendering strategy: using pre-rendered images when available to maximize refresh rate, and switching to real-time rendering when image authenticity is prioritized or pre-rendered images are not available. This dynamic adaptation resolves the contradiction by flexibly balancing between refresh rate and image quality based on current conditions.
2Productivity
If GPU rendering is performed continuously to maintain high frame rates, then display refresh rate is improved, but power consumption increases
Solution Approach 1:
The patent applies periodic action by alternating between rendering operations and idle states. Instead of continuous rendering, the system performs rendering during identified idle periods when the GPU is otherwise unused. This periodic utilization of idle GPU time achieves frame rate improvement without requiring continuous high-power rendering operations, thereby resolving the contradiction between productivity and energy consumption by exploiting otherwise wasted idle periods.
Solution Approach 2:
The patent implements self-service by having the system utilize its own idle resources for rendering purposes. The GPU's idle time, which would otherwise be wasted, is automatically exploited to perform rendering operations. The system monitors its own state and self-manages the rendering schedule based on idle detection, eliminating the need for external power-intensive continuous rendering while maintaining high frame rates through efficient use of available resources.
Data Source
AI summary
The present disclosure provides a method, apparatus and device for rendering virtual reality scenes. The method includes: obtaining a virtual reality scene and determining whether the virtual reality scene is in a rendering idle state; if the virtual reality scene is in the rendering idle state, performing image rendering on the virtual reality scene to generate a display image and store a correspondence between the display image and a display area; and obtaining a target area to be displayed of the virtual reality scene, calling a target display image corresponding to the target area according to the correspondence and displaying the target display image.