Ray Tracing With Full-View Depth Maps for Off-Screen Surfaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional screen space ray tracing methods lack depth information outside the screen space, leading to inaccurate ray tracing results and limitations in obtaining valid surfaces.
Innovation Solution
A ray tracing method that utilizes an information map with full-view depth information of a virtual scene, projecting tracing rays from the virtual scene to the information map to extend ray tracing beyond the screen space, allowing for accurate tracing of both on-screen and off-screen objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If screen space ray tracing method is used, then ray tracing can be performed in real-time, but depth information outside screen space is lost and ray tracing accuracy deteriorates
Solution Approach 1:
The patent extends ray tracing from 2D screen space to 3D world space by introducing an information map that stores depth information for the entire virtual scene. This dimensional expansion allows rays to be traced beyond the screen boundaries while maintaining real-time performance through efficient depth querying on the information map.
Solution Approach 2:
The patent pre-computes and stores depth information for all objects in the virtual scene onto an information map before ray tracing execution. This preliminary action enables rapid depth comparison during ray tracing without real-time computation overhead, resolving the contradiction between speed and accuracy.
2Device complexity
If ray tracing is limited to screen space, then processing complexity is reduced, but the ability to detect off-screen surfaces is lost
Solution Approach 1:
The patent introduces an information map as an intermediary data structure that bridges screen space and world space. This information map stores pre-computed depth information and serves as a mediator that enables reliable off-screen surface detection without requiring complex real-time 3D scene analysis during ray tracing.
3Quantity of substance
If depth information is only stored for screen space, then memory usage is reduced, but ray tracing cannot determine intersections with off-screen objects
Solution Approach 1:
The patent extracts depth information from the full 3D scene and stores it in a compressed 2D information map structure. This extraction approach retains essential depth data for memory efficiency while enabling complete scene coverage for accurate ray tracing, resolving the contradiction between memory usage and information completeness.
Data Source
Figure 1a~1b
Figure 2~3b
Figure 3c~3d
AI summary
The present application discloses a ray tracing method and apparatus, a computer device, a storage medium, and a computer program product. The method comprises: acquiring an information map of a virtual scene, wherein full-view depth information of the virtual scene is recorded in the information map, and the full-view depth information comprises a depth value of each virtual object in the virtual scene; in the virtual scene and on the basis of a reference point, generating a tracing ray used for ray tracing, wherein the reference point is a position where the virtual object having a ray tracing requirement is located in the virtual scene; projecting the tracing ray from the virtual scene to the information map to obtain a travel path of the tracing ray on the information map; and on the information map, performing ray tracing along the travel path to obtain a ray tracing result.