Point Cloud Visualization Using Normal-Based Occlusion Culling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for visualizing large-scale point clouds face issues with invalid rendering of occluded points and resource wastage due to the inability to efficiently manage and display complex point cloud data, leading to suboptimal visualization effects.
Innovation Solution
A method involving the construction of a balanced octree structure and normal level-of-detail (LOD) visualization nodes, combined with a node scheduling strategy that utilizes separate reading and rendering threads to optimize the visualization process based on the viewpoint and normal information of the point cloud.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If partitioned index is used to solve the problem of reading point clouds into memory, then the point cloud can be divided into tiles for rendering, but invalid rendering such as occluded point clouds and back-face point clouds still occurs, causing waste of computer rendering resources
Solution Approach 1:
The patent performs preliminary classification of point cloud nodes into visible and invisible categories before rendering by comparing node normals with view directions. This preliminary action identifies and marks invisible points (occluded and back-face) in advance, so they can be skipped during actual rendering, avoiding wasted rendering resources while maintaining rendering efficiency
Solution Approach 2:
The patent applies different rendering treatments to different regions of the point cloud based on their visibility characteristics. By calculating view directions and normals locally for each node, the system selectively renders only visible points while skipping invisible ones, creating a localized quality optimization that reduces overall rendering load without affecting visible area quality
2Loss of information
If large-scale point cloud data is visualized, then comprehensive 3D information can be displayed, but it consumes a lot of computer resources and affects visualization effect
Solution Approach 1:
The patent segments the large-scale point cloud into a hierarchical octree structure with multiple levels of detail. This segmentation divides the massive point cloud into manageable nodes that can be independently processed and rendered at appropriate levels, reducing memory consumption and computational load while preserving the complete 3D information through the hierarchical structure
Solution Approach 2:
The patent implements dynamic level-of-detail (LOD) rendering where the rendering resolution adapts based on the viewer's distance and perspective. Points are rendered at different levels of detail dynamically, with higher detail for nearby visible points and lower detail or aggregation for distant or occluded points, optimizing resource usage while maintaining information completeness
Data Source
AI summary
A method for visualizing a large-scale point cloud based on normal, including: (S1) according to a spatial structure of a point cloud data, constructing a balanced octree structure of a node point cloud; (S2) according to the balanced octree structure and normal information of a point cloud, constructing an octree structure with the normal information; and constructing a normal level-of-detail (LOD) visualization node through downsampling; and (S3) determining a node scheduling strategy according to a relationship between a viewpoint, a viewing frustum and a normal of a render node; and respectively calling a reading thread and a rendering thread to simultaneously perform reading and rendering according to the node scheduling strategy.