Real-Time 3D Scene Rendering via Spatial Grouping and Illumination Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ray tracing techniques for rendering 3D scenes in augmented reality applications are computationally expensive and memory-intensive, making them difficult to implement on devices like mobile phones, tablets, and AR glasses for real-time display and video capture.
Innovation Solution
A method for rendering a 3D scene of a video that involves identifying parameters for objects, grouping them based on these parameters, determining spatial boundaries and illumination contributions for intermediate structures, and rendering the scene using these determined boundaries and contributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional ray tracing techniques are used for rendering 3D scenes in augmented reality, then high quality lighting effects (reflections, refractions, shadows) are achieved, but computational cost and memory usage become excessively high, making real-time implementation on mobile devices difficult
Solution Approach 1:
The patent segments the 3D scene into multiple groups based on object parameters (such as material properties, geometric complexity, and spatial location). Each group is processed independently with appropriate rendering techniques, allowing high-quality ray tracing for critical objects while using simpler methods for less important objects, thus achieving real-time performance without sacrificing overall visual quality
Solution Approach 2:
The patent applies different rendering quality levels to different regions and objects within the 3D scene. Important objects with complex lighting interactions receive high-quality ray tracing, while background or less significant objects are rendered with simplified techniques. This local differentiation maintains lighting accuracy where needed while reducing overall computational burden for real-time operation
2Manufacturing precision
If traditional ray tracing techniques are used for rendering 3D scenes in augmented reality, then accurate lighting effects are obtained, but memory consumption becomes excessively high, making implementation on mobile devices difficult
Solution Approach 1:
The patent divides the scene into discrete object groups and processes them independently, reducing the memory footprint by avoiding the need to store complete ray tracing data structures for all objects simultaneously. Only essential lighting information is retained for each group, significantly reducing memory requirements while maintaining rendering accuracy
Solution Approach 2:
The patent uses simplified representations or proxies for complex objects when computing lighting effects. Instead of storing complete geometric and material data for all objects, the system creates simplified models that capture essential lighting interaction characteristics, reducing memory consumption while preserving rendering quality for the final output
Data Source
AI summary
A method and a system for rendering a 3D scene are disclosed. A set of parameters for objects in the 3D scene are identified. The objects are grouped based on the identified set of parameters and spatial boundaries for an intermediate structure group of objects based on an object type is determined. An illumination contribution for each plane of the spatial boundary of the intermediate structure is determined for the objects in the 3D scene. The 3D scene is then rendered based on the determined spatial boundary and the illumination contribution of the intermediate structure for each object in the scene.