Point Cloud Reconstruction with Trisoup Continuity at Slice Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In geometry-based point cloud compression, trisoup-based encoding methods result in broken trisoup structures at slice partitioning, leading to discontinuous surfaces and interstices during point cloud model reconstruction, affecting the quality of the reconstructed model.
Innovation Solution
The method involves determining the i-th segmentation position and vertex position within each three-dimensional slice, constructing a trisoup based on these positions, and using vertex information from intersections between slices to ensure continuous surface fitting during encoding and decoding, thereby eliminating interstices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If slice partitioning is performed along the longest edge in trisoup-based encoding, then encoding efficiency is improved, but the trisoup structure becomes broken at slice boundaries, causing discontinuous surfaces and interstices in the reconstructed point cloud model
Solution Approach 1:
The patent divides the point cloud space into multiple slices along the longest edge for independent parallel encoding, improving efficiency. However, it also segments the affected blocks at slice boundaries into sub-blocks to preserve trisoup continuity, resolving the contradiction between encoding efficiency and reconstruction quality.
Solution Approach 2:
The patent applies different processing strategies to different regions: blocks completely within slices use standard trisoup fitting, while blocks at slice boundaries use special handling with sub-block segmentation. This local differentiation maintains high reconstruction quality at critical boundary regions while preserving overall encoding efficiency.
2Speed
If independent parallel encoding of slices is used, then encoding speed is improved, but vertex coordinates on edges belong to two neighbouring slices, breaking the trisoup structure
Solution Approach 1:
The patent identifies blocks affected by slice boundaries in advance and performs preliminary segmentation into sub-blocks before trisoup fitting. This preliminary action ensures that vertex coordinates are properly assigned and trisoup structure continuity is maintained, enabling subsequent parallel encoding without structural breaks.
Solution Approach 2:
The patent introduces sub-blocks as intermediary structures at slice boundaries, which mediate between neighbouring slices. These sub-blocks contain the shared vertex coordinates and enable continuous trisoup fitting across slice boundaries, allowing independent parallel encoding while maintaining structural stability.
3Device complexity
If sampling is performed on broken trisoup at slice boundaries, then encoding complexity is reduced, but no sampling points are obtained on discontinuous parts, creating interstices
Solution Approach 1:
The patent performs preliminary segmentation of affected blocks into sub-blocks and identifies sampling points on the continuous trisoup structure before encoding. This ensures that sampling points are obtained on all necessary surfaces including previously discontinuous parts, eliminating interstices while maintaining manageable encoding complexity.
Data Source
AI summary
A point cloud model reconstruction method, an encoder, and a decoder are provided. The method includes the following. An i-th segmentation position of an i-th three-dimensional slice on a longest edge in a point cloud space is obtained, where 1≤i≤n−1 and n is a total number of three-dimensional slices to-be-processed. An i-th cube block set nearest to the i-th segmentation position in the i-th three-dimensional slice is determined based on the i-th segmentation position and a preset cube block length. An i-th vertex position with a cube block in the i-th cube block set is determined along the longest edge, where the i-th vertex position represents a position of a vertex on an edge which is away from a start position of the cube block by the preset cube block length. A point cloud model is reconstructed based on the i-th vertex position.