Subjective Video Quality Evaluation via Adaptive Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for assessing video quality rely heavily on objective metrics, which fail to reliably capture subjective quality, especially across various artifact types and extremities, leading to the need for a fast, reliable, and cost-effective subjective quality assessment process.
Innovation Solution
A computerized system comprising a server and client application that facilitates subjective video quality assessment by presenting users with pairs of video clips, allowing them to provide feedback on the relative quality, with the server analyzing the feedback to provide subjective quality evaluations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If objective quality metrics (such as PSNR, SSIM, VMAF) are used to assess video quality, then the assessment process is fast, low-cost, and repeatable, but the results cannot reliably indicate subjective quality across various artifact types and extremities
Solution Approach 1:
The patent introduces an intermediary mapping model that translates objective metric results into subjective quality estimates. This mapping model acts as a mediator between the automated objective assessment system and human subjective perception, allowing the system to leverage the speed of objective metrics while compensating for their inability to accurately reflect subjective quality across different artifact types through learned transformation relationships.
Solution Approach 2:
The system dynamically adjusts assessment parameters based on the type of video artifacts detected. When specific artifact types are identified through objective metrics, the system modifies its assessment strategy by applying artifact-specific mapping models or adjusting weighting factors, thereby adapting the generic objective metric framework to handle diverse degradation scenarios more accurately.
2Measurement precision
If standardized subjective testing methods (such as ITU-R BT.500) are used to assess video quality, then the results reliably capture subjective quality perception, but the assessment process becomes complex, time-consuming, and costly
Solution Approach 1:
The patent extracts the essential function of subjective assessment from the complex standardized testing framework. Instead of implementing the full ITU-R BT.500 methodology with its multiple test methods, observer training requirements, and controlled viewing conditions, the system extracts only the core objective of capturing subjective quality perception and achieves it through simplified automated metric-based assessment with adaptive mapping.
Solution Approach 2:
The system creates a computational model that copies the behavior and outcomes of standardized subjective testing without requiring actual human observers. By training mapping models on datasets that include both objective metrics and subjective quality ratings from standardized tests, the system learns to replicate the results of complex subjective assessments through simpler automated processing.
3Measurement precision
If standardized subjective testing methods are used to ensure reliable quality assessment, then the results are accurate, but the assessment process requires significant time and computational resources
Solution Approach 1:
The system performs preliminary actions by pre-training mapping models on comprehensive datasets that encompass various video artifact types and quality levels before actual assessment. This preprocessing step creates ready-to-use transformation models that can quickly estimate subjective quality without requiring time-consuming real-time analysis or human observer involvement during the actual assessment process.
Solution Approach 2:
The patent replaces the mechanical system of human observers physically viewing and rating video content with an automated computational system. By substituting human visual processing and subjective judgment with algorithmic analysis of objective metrics through trained mapping models, the system eliminates the time required for human observation while maintaining assessment reliability through data-driven transformations.
Data Source
AI summary
There are provided computerized systems and methods for video quality assessment, the system including a server including a database configured to store pointers to a plurality of video clip pairs and a processor operatively connected thereto, configured to store pointers to a plurality of video clip pairs to be used in one or more test sessions performed by one or more users. The processor is configured to create, for each test session performed by a respective user: a test set, a display order and display positions of the each video in each video clip pairs in the set, and is further configured to send the one or more test sets to the one or more users for assessing quality and to receive feedback regarding the assessed quality, wherein the feedback is usable for providing subjective quality evaluation of the plurality of video clip pairs.