3D Point Cloud Decoding with Extended-Triangle Voxelization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud compression technologies struggle to achieve high accuracy and efficient compression of dense and dynamic 3D point clouds, particularly in AR/VR applications, due to limitations in existing codecs and the need for real-time transmission with low bitrates.
Innovation Solution
A method for decoding and encoding 3D point clouds using an octree structure and TriSoup encoding, where triangles are extended along their sides during voxelization to enhance accuracy, and the Möller-Trumbore algorithm is used with relaxed convex hull requirements to capture additional points, improving reconstruction quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional voxelization methods are used for decoding point cloud geometry, then the decoding process is simple, but the reconstruction accuracy is insufficient and sampling errors occur
Solution Approach 1:
The patent applies preliminary action by extending triangles along their sides before the voxelization process. Specifically, the method determines extended triangles by connecting vertices on edges of cuboids and extends these triangles along their sides to create a larger coverage area. This preliminary extension ensures that additional points that would otherwise be missed during standard voxelization are captured, thereby improving reconstruction accuracy without significantly increasing decoding complexity
Solution Approach 2:
The patent implements dimensionality change by transitioning from standard triangle voxelization to extended triangle voxelization. The extension along triangle sides effectively adds spatial coverage in additional dimensional directions, allowing the voxelization process to capture points that lie outside the original triangle boundaries but within the extended region, thus improving measurement precision
2Productivity
If compression is applied to reduce bitrate for transmission, then transmission efficiency improves, but reconstruction quality deteriorates
Solution Approach 1:
The patent uses preliminary action by pre-extending triangles along their sides before voxelization. This extension creates a larger sampling area that captures additional geometric points that would otherwise be lost during compression. By performing this extension beforehand, the method ensures that even at low bitrates, the reconstructed point cloud maintains higher geometric fidelity and captures more accurate surface details
Solution Approach 2:
The patent applies parameter changes by modifying the triangle geometry parameters through extension along their sides. This parameter modification changes the spatial coverage of triangles from their original boundaries to extended boundaries, allowing the voxelization process to sample a broader region and capture additional points, thereby improving reconstruction quality without requiring increased bitrate
3Measurement precision
If standard triangle connection methods are used, then the encoding process is simple, but additional points are missed and accuracy is reduced
Solution Approach 1:
The patent implements preliminary action by determining and extending triangles along their sides before the voxelization step. The method connects vertices on edges of cuboids to form triangles, then extends these triangles along their sides to create extended triangles. This preliminary extension ensures that the encoding process captures additional geometric points that would be missed by standard triangle connection methods, improving point capture accuracy
Solution Approach 2:
The patent applies segmentation by dividing the encoding process into distinct stages: determining vertices on edges, connecting vertices to form triangles, extending triangles along their sides, and finally performing voxelization. This segmented approach allows each step to be optimized independently, capturing additional points through the extension step while maintaining manageable encoding complexity
Data Source
AI summary
Method for decoding, from a bitstream, the geometry of a 3D point cloud by decoding information constituting a geometrical representation of the geometry, preferably implemented in a decoder, including: receiving and decoding a bitstream, wherein the bitstream contains octree information including information about octree structure of the volume of the point cloud and vertex information including information about vertex presence and position of a vertex on edges of the cuboids; determining triangles by connecting the vertices of one cuboid of the octree structure; voxelization of the triangles to determine points of the point cloud, wherein at least one triangle is extended at along at least one side for voxelization.