Neural Network Video Quality Assessment Using Frame ID Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video quality assessment methods for video-based applications in cloud environments use inflexible metrics that may not accurately reflect video quality, making it challenging to optimize cloud resources while maintaining user experience.
Innovation Solution
A system and method that inserts frame identifiers into video content and uses a text recognition neural network to calculate the frame per second (FPS) metric, allowing for objective video quality assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional inflexible metrics are used to quantify video quality, then the assessment process is simple, but the accuracy of video quality reflection is poor
Solution Approach 1:
The patent changes the parameters used for video quality assessment from traditional inflexible metrics to flexible neural network-based metrics. The system inserts frame identifiers into video content and uses a text recognition neural network to recognize these identifiers, calculating FPS metrics and assessing objective video quality. This parameter change enables more accurate video quality measurement while maintaining manageable system complexity through automated processing
Solution Approach 2:
The patent replaces traditional mechanical/video-processing-based quality assessment systems with a neural network-based system. Instead of using conventional video processing techniques, the system uses a text recognition neural network to identify frame identifiers and calculate quality metrics, substituting complex video analysis mechanics with AI-based pattern recognition and calculation
2Reliability
If cloud resources are optimized to reduce costs, then operational expenses decrease, but video quality for end-users may deteriorate
Solution Approach 1:
The patent implements a feedback mechanism where video quality is continuously monitored using neural network-based assessment. The system inserts frame identifiers, recognizes them via neural network, calculates FPS metrics, and assesses objective video quality. This feedback loop enables cloud providers to monitor actual video quality under different resource allocation scenarios, allowing them to optimize cloud resources while maintaining quality consistency through data-driven decision-making
Solution Approach 2:
The patent changes the approach to resource optimization by introducing neural network-based video quality assessment parameters. Instead of relying on traditional metrics, the system uses FPS calculations and objective quality assessments derived from neural network analysis of frame identifiers. This enables more precise control over the balance between cloud resource consumption and video quality maintenance
Data Source
AI summary
A system and method for assessing video quality of a video-based application inserts frame identifiers (IDs) into video content from the video-based application and recognizes the frame IDs from the video content using a text recognition neural network. Based on recognized frame IDs, a frame per second (FPS) metric of the video content is calculated. Based on the FPS metric of the video content, objective video quality of the video-based application is assessed.