Dynamic Point Cloud Processing with 2D Geometry Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently compressing dynamic point clouds while maintaining an acceptable quality of experience, as they often require significant bitrate or storage space, which is crucial for distributing immersive worlds and other applications.
Innovation Solution
A two-layer-based point cloud encoding structure is employed, comprising a base layer for lossy representation and an enhancement layer for higher quality, along with image-based encoding using existing video codecs to convert point cloud data into video sequences, and metadata for interpretation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dynamic point clouds are compressed using existing technologies, then quality of experience is maintained, but bitrate and storage space requirements become excessive
Solution Approach 1:
The patent divides the point cloud data into multiple layers (base layer and enhancement layers) with different quality levels. The base layer provides essential information at lower bitrate, while enhancement layers add incremental quality improvements. This segmentation allows receivers to select appropriate layers based on available bandwidth and storage, resolving the contradiction between quality and resource consumption.
Solution Approach 2:
The patent transforms point cloud data into alternative representations by changing parameters such as coordinate systems (e.g., transforming to spherical coordinates), sampling rates, and precision levels. This parameter transformation enables more efficient compression by matching the data representation to the specific application requirements, reducing bitrate while maintaining acceptable quality.
2Adaptability or versatility
If point cloud data is distributed to end-users, then immersive world distribution is enabled, but consumption of bitrate and storage space increases
Solution Approach 1:
The patent implements dynamic adaptation of point cloud data distribution by adjusting the level of detail, resolution, and quality based on receiver capabilities, network conditions, and storage availability. The system can dynamically select which layers to transmit and at what quality level, enabling versatile distribution across different platforms while optimizing bitrate consumption according to actual needs.
3Reliability
If static colored huge point clouds are used for culture heritage and topography, then preservation and visualization are achieved, but data size becomes unmanageable
Solution Approach 1:
The patent extracts and separates different attributes of point cloud data (geometry, color, texture, semantic information) into independent layers or representations. This extraction allows selective compression and transmission of only the essential preservation-critical attributes, reducing overall data size while maintaining reliability for heritage and topography applications.
Data Source
Figure 1~2
Figure 3
Figure 3a
AI summary
At least one embodiment relates to a method for smoothing (filtering) the geometry of a point cloud frame by performing an analysis and filtering of said geometry point cloud in a 2D space, without reconstruction of 3D samples in a 3D space, and by using a flexible filtering operator that, in addition to moving existing points, can also remove points or add new ones.