Adaptive Point Cloud Attribute Coding for Lower Residuals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for coding and decoding point cloud attribute information fail to consider the discontinuity of actual scenarios and correlations between different pieces of information, leading to low prediction accuracy, large predicted residuals, and frequent outliers.
Innovation Solution
Create an adaptive prediction list for point cloud attribute information by selecting points based on depth, spatial, and azimuth information, and design an entropy coding context model to optimize prediction modes and residuals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If attribute information is coded based on recreated geometric information using Morton code for neighboring search, then geometric structure is preserved, but prediction accuracy deteriorates due to scenario discontinuity
Solution Approach 1:
The patent applies dynamics by making the prediction list adaptive and dynamic rather than static. The prediction list is updated continuously during the coding process by selecting points based on depth, spatial, and azimuth information, allowing the system to adapt to scenario discontinuities and improve prediction accuracy for attribute information coding.
Solution Approach 2:
The patent changes multiple parameters simultaneously: it uses depth information, spatial position, and azimuth information as selection criteria for building the prediction list. This multi-parameter approach allows the system to account for scenario discontinuities and improve prediction accuracy by selecting more relevant neighboring points.
2Productivity
If conventional prediction methods are used without considering scenario discontinuity, then coding process is simple, but predicted residual increases leading to poor coding efficiency
Solution Approach 1:
The patent adds multiple dimensions to the prediction process by incorporating depth information, spatial position, and azimuth information as selection criteria. This multi-dimensional approach allows the system to better capture the structure of the point cloud and reduce predicted residuals, thereby improving coding efficiency without excessive complexity.
Solution Approach 2:
The patent implements feedback by continuously updating the prediction list during the coding process. Points are selected and added to the prediction list based on their attribute information and spatial relationships, creating a feedback loop that improves prediction accuracy for subsequent points and reduces overall predicted residuals.
3Measurement precision
If adaptive prediction list is created using depth, spatial, and azimuth information, then prediction accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-sorting and organizing points based on depth, spatial, and azimuth information before the actual attribute coding process. This preliminary organization allows the prediction list to be built more efficiently during coding, reducing the computational burden despite the multi-parameter selection criteria.
Data Source
AI summary
A method for coding includes: obtaining original point cloud data; creating an adaptive prediction list of the attribute information of the point cloud; selecting a prediction mode from the adaptive prediction list and predicting the attribute information of the point cloud, to obtain a predicted residual; and coding the prediction mode and the predicted residual, to obtain codestream information.