Video Quality Estimation via Bitstream Error Propagation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video quality estimation techniques face challenges in achieving accurate, real-time, and low-complexity assessments, with perceptual models being too complex and computational resource-intensive, while parametric models provide low estimation accuracy due to limited network protocol header information.
Innovation Solution
A method for estimating video quality by determining error occurrence and propagation in a video bitstream, calculating visual degradation using a scoring scheme, and aggregating penalties to provide an overall quality score, which can be implemented in various nodes or terminals without decoding the bitstream, utilizing parameters from the video coding layer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If perceptual models are used for video quality estimation, then estimation accuracy is improved, but device complexity and computational resource consumption increase
Solution Approach 1:
The patent extracts only the essential features needed for quality assessment from the video bitstream, specifically focusing on error occurrence and propagation characteristics rather than processing the entire decoded video content. This extraction approach maintains accuracy while reducing computational burden by taking out only the critical quality-determining elements.
Solution Approach 2:
The patent segments the video quality assessment into distinct components: error detection, error propagation analysis, and quality scoring. By dividing the assessment process into manageable segments that operate on specific bitstream parameters rather than full video content, the model achieves perceptual-level accuracy with reduced complexity.
2Measurement precision
If perceptual models are used for video quality estimation, then estimation accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary analysis of error occurrence and propagation patterns directly from the bitstream before full quality assessment is needed. By pre-identifying error locations and propagation paths, the model prepares quality indicators in advance, reducing the time required for final quality determination while maintaining accuracy.
3Device complexity
If parametric models are used for video quality estimation, then device complexity is reduced, but estimation accuracy deteriorates
Solution Approach 1:
The patent changes the parameters used in parametric models from basic network protocol headers to include detailed error propagation characteristics and visual degradation metrics derived from video coding layer parameters. This parameter enhancement allows the simplified parametric model to achieve accuracy comparable to perceptual models by focusing on the most quality-relevant parameters.
4Use of energy by moving object
If parametric models are used for video quality estimation, then computational resource consumption is reduced, but estimation accuracy deteriorates
Solution Approach 1:
The patent applies local quality assessment by analyzing error propagation and visual degradation at specific locations and time points in the video bitstream rather than requiring global video content analysis. This localized approach consumes fewer computational resources while maintaining accuracy by focusing only on the regions and parameters that most affect quality.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
An objective video quality estimation technique is disclosed. The technique may be based on a video bitstream model, using parameters taken from the video coding layer of the bitstream for estimating the quality of the video. The technique can be implemented as a method, a computer program, a computer program product, a device, or any one of a server node, a client terminal and a network node comprising the device. As a method embodiment, the technique comprises receiving a video bitstream comprising a series of picture frames; determining an error occurrence in a picture frame of the video bitstream; determining at least one of a temporal propagation and a spatial propagation of the error; and estimating the quality of the video bitstream based on result of the determination.