Precomputed Cell Generation for Virtual Scene Navigation Meshes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing precomputed cell generation and display methods struggle to accurately cover all reachable regions in virtual scenes, especially when full-view spaces are not limited by the navigation mesh, leading to a low matching degree between precomputed cells and the virtual scene.
Innovation Solution
The method involves displaying and generating precomputed cells in response to view space division events in virtual scenes. It differentiates between limited-view and full-view spaces, using navigation meshes in limited-view spaces and static meshes in full-view spaces to extend precomputed cells in the standing direction of virtual characters, thereby covering these meshes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a navigation mesh is used to represent reachable regions in full-view spaces, then the precomputed cells can be generated based on character movement, but the matching degree between precomputed cells and virtual scene is not high because full-view spaces are not limited by the navigation mesh
Solution Approach 1:
The virtual scene is divided into limited-view spaces and full-view spaces based on whether the navigation mesh limits the reachable region. Different mesh types are applied to different space types: navigation meshes for limited-view spaces and static meshes for full-view spaces, ensuring precise matching while maintaining comprehensive coverage
Solution Approach 2:
Different mesh representations are used in different regions: navigation meshes in limited-view spaces where character movement is constrained, and static meshes in full-view spaces where the character can reach any position. This local differentiation optimizes the matching degree in each region while maintaining overall coverage
2Productivity
If precomputed cells are generated based on navigation mesh only, then the generation process is simple, but the coverage of reachable regions is incomplete when full-view spaces exist
Solution Approach 1:
The view space is pre-divided into limited-view and full-view spaces before generating precomputed cells. This preliminary classification allows the system to automatically select the appropriate mesh type (navigation mesh or static mesh) for each space type, ensuring complete coverage without requiring complex real-time calculations
Solution Approach 2:
The system handles both limited-view and full-view spaces using a unified precomputed cell generation framework. By supporting multiple mesh types (navigation meshes and static meshes) within the same system, it achieves universal applicability across different scene types while maintaining consistent processing logic
Data Source
AI summary
This application relates to a precomputed cell display method and apparatus, a computer device, a computer-readable storage medium, and a computer program product. The method includes: displaying, in response to a view space division event triggered for a virtual scene in which a virtual character is located, a limited-view space and a full-view space that are obtained through division in the virtual scene; displaying, corresponding to a navigation mesh of a first space in the limited-view space, a plurality of precomputed cells that extend in a standing direction of the virtual character and that cover the navigation mesh; and displaying, corresponding to a static mesh of a second space in the full-view space, a plurality of precomputed cells that extend in the standing direction of the virtual character and that cover the static mesh of the second space.