Parallel Mesh Edge Decimation Without Partition Interface Conflicts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for edge decimation in graphics computing systems are limited by partition size dependencies and often result in issues at the interface of partitions, leading to suboptimal performance and quality.
Innovation Solution
A method for edge decimation that generates edge costs for each edge in a mesh, selects collapse candidates based on these costs, propagates costs to neighboring triangles, and collapses edges while restricting ineligible edges in the neighborhood, allowing multiple edges to be collapsed in the same iteration without intersecting neighborhoods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the mesh is partitioned into voxels for parallel processing, then processing speed is improved, but problems occur at partition interfaces and performance becomes dependent on partition size
Solution Approach 1:
The patent divides the mesh into multiple independent batches, where each batch contains a subset of edges that can be processed in parallel. This segmentation allows simultaneous processing of multiple edges without the interface problems that occur with spatial voxel partitioning, as each batch is independently selectable based on collapse cost criteria.
Solution Approach 2:
The patent dynamically selects edges for collapse based on their cost values, allowing the processing to adapt to the actual mesh structure rather than being constrained by fixed spatial partitions. Edges are grouped into batches dynamically based on which edges have the lowest collapse costs, enabling flexible parallel processing that avoids static partition interface issues.
2Manufacturing precision
If only one edge is collapsed per partition per iteration, then quality is maintained, but the number of collapse operations per iteration is limited
Solution Approach 1:
The patent merges multiple edge collapse operations into a single iteration by selecting multiple independent edges based on their collapse costs. Instead of limiting to one edge per partition, the system identifies and processes multiple edges across different batches simultaneously, each meeting the quality criteria for collapse, thereby increasing throughput while maintaining quality standards.
Solution Approach 2:
The patent performs more collapse operations per iteration than traditional methods by processing multiple batches in parallel. Each batch contains edges that satisfy the collapse quality criteria, and by executing multiple batches simultaneously, the system performs an excessive number of collapses relative to traditional sequential methods, then refines the result in subsequent iterations.
3Productivity
If multiple edges are collapsed in parallel, then productivity is improved, but conflicts arise when edges have intersecting neighborhoods
Solution Approach 1:
The patent segments the set of candidate edges into multiple batches, where each batch contains edges that are independent of each other (no shared vertices or triangles). This segmentation eliminates conflicts between parallel collapse operations while still allowing multiple edges to be processed simultaneously, as each batch can be collapsed independently without affecting the others.
Solution Approach 2:
The patent performs preliminary classification of edges into batches based on their collapse costs and independence criteria before executing parallel collapse operations. By pre-organizing edges into conflict-free batches, the system ensures that subsequent parallel operations will not have conflicts, maintaining reliability while achieving high productivity.
Data Source
AI summary
Various embodiments include techniques for performing parallel edge decimation on a high resolution mesh by collapsing multiple edges in parallel by blocking only the neighbor edges of the edges selected as collapse candidates. Effectively, the disclosed techniques dynamically partition the mesh into small partitions around the collapse candidates. In this manner, the techniques identify all the edges that may be independently collapsed in a single, now parallel, iteration. Edge decimation may be performed so that certain computational geometry techniques can be efficiently applied to a simpler mesh. In so doing, the disclosed techniques preserve the history of how the edge decimation process displaces the vertices of the original mesh to generate the simplified mesh. As a result, the results of the computational geometry techniques as applied to the simplified mesh can be propagated back to the original mesh.