Point Cloud Coding With Multiple Reference Frames
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for compressing point clouds face challenges in efficiently coding motion vectors, particularly in dynamic point clouds where temporal correlations are high.
Innovation Solution
The proposed solution involves using multiple frames of reference and determining a transform between them to efficiently encode and decode point clouds. This approach segments the volumetric space into regions associated with each frame of reference, allowing for improved prediction and context-based entropy encoding of occupancy data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional single frame of reference is used for point cloud coding, then motion vector search is straightforward, but coding efficiency is poor and complexity is high
Solution Approach 1:
The patent divides the volumetric space into multiple regions, each associated with a different frame of reference. This segmentation allows motion vectors to be computed relative to the most appropriate reference frame for each region, improving coding efficiency while managing complexity through localized processing
Solution Approach 2:
The patent introduces multiple frames of reference as an additional dimension to the traditional single-reference approach. By transforming point cloud data into multiple reference frames and selecting the optimal one for each region, the system achieves better compression efficiency without proportionally increasing overall complexity
2Loss of information
If multiple frames of reference are used, then compression efficiency is improved, but computational complexity increases
Solution Approach 1:
Different regions of the volumetric space are associated with different frames of reference based on local characteristics. This local quality approach ensures that transform computations are performed only where beneficial, improving compression efficiency while minimizing unnecessary computational complexity
Solution Approach 2:
The patent applies multiple frames of reference partially - only to regions where they provide benefit. Not all regions require multiple reference frames, and the system selectively applies transforms based on local needs, achieving improved compression without excessive computational overhead
3Measurement precision
If motion vectors are coded with high precision, then prediction accuracy is improved, but bit rate increases
Solution Approach 1:
The patent performs preliminary transformation of point cloud data into multiple reference frames before motion vector computation. This preliminary action creates a favorable basis for finding smaller, more compact motion vectors, improving prediction accuracy while reducing the bit rate required to encode them
Data Source
AI summary
Methods and devices for encoding a point cloud. More than one frame of reference is identified and a transform defines the relative motion of a second frame of reference to a first frame of reference. The space is segmented into regions and each region is associated with one of the frames of reference. Local motion vectors within a region are expressed relative to the frame of reference associated with that region. Occupancy of the bitstream is entropy encoded based on predictions determined using the location motion vectors and the transform associated with the attached frame of reference.