Multi-Resolution Video Quality Evaluation Using Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for evaluating video quality are inefficient and subjective, making it difficult to accurately predict the quality perceived by viewers, especially when video sequences are transmitted with reduced resolution.
Innovation Solution
An apparatus and method that includes an encoder unit to reduce and encode video sequences, a decoder unit to interpolate and decode them, and a video quality index calculator to objectively evaluate video quality using multi-resolution comparisons.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If video sequences are transmitted with reduced resolution to save bandwidth and storage, then transmission efficiency is improved, but video quality perceived by viewers deteriorates
Solution Approach 1:
The patent segments the video quality evaluation into multiple resolution levels. Instead of evaluating only at the original high resolution, the method divides the evaluation into different resolution stages (original resolution and reduced resolution), allowing separate assessment of each transformation step and their combined impact on perceived quality
Solution Approach 2:
The patent adds a resolution dimension to the quality evaluation process. By introducing multi-resolution analysis, the system evaluates video quality not just in terms of traditional metrics but across different resolution dimensions, capturing how quality degrades through resolution reduction and interpolation
2Measurement precision
If subjective evaluation methods are used to assess video quality, then evaluation accuracy is improved, but evaluation efficiency and time consumption deteriorate
Solution Approach 1:
The patent replaces the mechanical system of human subjective evaluation with an automated objective evaluation system. The method uses algorithmic quality metrics that automatically assess video quality at multiple resolutions, eliminating the need for human evaluators while maintaining evaluation accuracy through sophisticated comparison algorithms
Solution Approach 2:
The evaluation system performs self-assessment by automatically comparing encoded video sequences against reference sequences at multiple resolutions. The system generates its own evaluation metrics without external human intervention, enabling continuous automated quality monitoring
3Device complexity
If single-resolution evaluation is performed for simplicity, then evaluation complexity is reduced, but evaluation accuracy for multi-resolution videos deteriorates
Solution Approach 1:
The patent segments the quality evaluation process into distinct resolution-specific evaluation stages. Each resolution level (original and reduced) undergoes separate quality assessment, and the results are combined to provide a comprehensive quality metric that reflects the entire processing chain
Data Source
AI summary
A method and apparatus for evaluating video quality are disclosed. An apparatus for evaluating video quality according to an embodiment of the invention includes: an encoder unit that converts a first-resolution original video sequence into a second-resolution original video sequence and encodes the second-resolution original video sequence; decoder unit that generates a second-resolution decoded video sequence by decoding the encoded second-resolution video sequence and converts the second-resolution decoded video sequence into a first-resolution decoded video sequence by interpolating the second-resolution decoded video sequence to the first resolution; a video quality index calculator unit that calculates a first video quality index by comparing the first-resolution original video sequence with the first-resolution decoded video sequence and calculates a second video quality index by comparing the second-resolution original video sequence with the second-resolution decoded video sequence; and a video quality evaluator unit that calculates a final video quality index by using the first video quality index and the second video quality index.