Vertex Occlusion Encoding for Secondary Light Rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for rendering secondary light occlusion in game development and animation consume excessive device performance, leading to high bandwidth usage and unnatural light effects, particularly on mobile devices, due to frame-by-frame calculation and reliance on render textures.
Innovation Solution
A method that detects vertex occlusion information in advance, encodes it, and decodes it based on secondary light direction information to generate occlusion rendering data, reducing the need for frame-by-frame texture reading and improving performance efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frame-by-frame calculation is performed for secondary light occlusion using render textures, then occlusion accuracy is improved, but device performance consumption increases excessively
Solution Approach 1:
The patent performs light occlusion detection on vertexes in advance before rendering, encoding the detection results and storing them in vertex data. This preliminary action eliminates the need for frame-by-frame calculation during rendering, significantly reducing real-time computational overhead while maintaining occlusion accuracy.
Solution Approach 2:
The patent creates an encoded copy of occlusion information that is stored with vertex data and reused during rendering. Instead of recalculating occlusion for each frame, the system uses the pre-computed encoded occlusion information, reducing redundant calculations and improving performance.
2Measurement precision
If render textures are used for each secondary light to store minimum depth information, then occlusion detection accuracy is improved, but bandwidth usage increases due to excessive texture reading
Solution Approach 1:
The patent extracts occlusion information from the render texture reading process and embeds it directly into vertex data during preprocessing. By taking out the essential occlusion information and storing it with vertex attributes, the system eliminates the need to read multiple render textures during rendering, significantly reducing bandwidth consumption.
Solution Approach 2:
The patent moves occlusion information from the texture memory dimension to the vertex data dimension. By encoding occlusion information in vertex data that is already loaded into GPU memory for rendering, the system avoids additional texture memory reads and reduces bandwidth requirements.
3Illumination intensity
If multiple secondary lights are rendered with occlusion calculation, then visual realism is improved, but power consumption increases due to high bandwidth usage
Solution Approach 1:
The patent performs occlusion detection and encoding for all secondary lights in advance during preprocessing. This preliminary action allows the rendering stage to use pre-computed occlusion information without intensive calculations, reducing real-time power consumption while maintaining visual realism for multiple lights.
Solution Approach 2:
The patent uses encoded copies of occlusion information stored in vertex data for each secondary light. During rendering, the system retrieves and uses these pre-computed copies instead of performing new occlusion calculations, reducing the computational and bandwidth overhead that would otherwise increase power consumption.
Data Source
AI summary
This application discloses a secondary light occlusion rendering method performed by a computer device. The method includes: obtaining direction information of a secondary light in an environment including a virtual object and data of a plurality of vertexes on a model of the virtual object; decoding the encoding results based on the direction information to obtain decoded vertex occlusion information; and obtaining secondary light occlusion rendering data of the virtual object based on the decoded vertex occlusion information respectively corresponding to the plurality of vertexes and light intensity information of the secondary light. In this way, textures do not need to be read frame by frame, thereby reducing consumption of performance of a computing device. The rendering solution is based on vertexes on a model and is not affected by a change in a movement direction of a virtual object.