Point Cloud Quality Assessment Using Unified Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Video-based Point Cloud Compression (V-PCC) technologies cannot accurately reflect the subjective quality of point clouds as they independently calculate geometry and color Peak Signal to Noise Ratio (PSNR), failing to account for the simultaneous visual impact on human eye experience.
Innovation Solution
A point cloud quality assessment method that determines a subjective quality measurement value using a quality assessment model, based on feature parameters extracted from the point cloud, simplifying computational complexity and improving accuracy by eliminating the need for distortion point pairs and original point clouds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If geometry PSNR and color PSNR are calculated independently using PC_error technology, then the calculation process is simple and straightforward, but the assessment accuracy of subjective quality is insufficient because it cannot reflect the simultaneous visual impact on human eye experience
Solution Approach 1:
The patent combines geometry PSNR and color PSNR into a unified quality assessment model that processes both geometric and color features simultaneously. The model integrates multiple feature parameters (geometry PSNR, color PSNR, and their interaction terms) to produce a single comprehensive quality score, reflecting the simultaneous visual impact on human eye experience while maintaining computational efficiency through a unified mathematical framework.
2Measurement precision
If a comprehensive quality assessment model is used to reflect simultaneous geometry and color impact, then the subjective quality assessment accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent transforms the quality assessment by changing the parameters from separate independent calculations to a unified model that uses feature parameters extracted from the point cloud. The model employs a mathematical formulation that combines geometry PSNR (α·PSNR_geo) and color PSNR (β·PSNR_color) with interaction terms, where α and β are weighting coefficients. This parameter transformation enables the model to capture the simultaneous visual impact while maintaining computational efficiency through a closed-form solution.
Data Source
AI summary
Disclosed are a point cloud quality assessment method, an encoder and a decoder. The method comprises: decoding a bitstream to acquire a feature parameter of a point cloud to be assessed; determining a model parameter of a quality assessment model; and according to the model parameter and the feature parameter of the point cloud, determining a subjective quality measurement value of the point cloud using the quality assessment model.