Video Quality Assessment Using Blur Segmentation and Dual Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video quality assessment methods, particularly the no-reference quality assessment method, face challenges in accurately evaluating the quality of video frames, especially in distinguishing between fully-blurred and partially-blurred frames, and in considering the characteristics of sub-areas within a frame, which affects the reliability and efficiency of quality evaluation.
Innovation Solution
A method and apparatus that identify whether a frame is fully-blurred or partially-blurred based on its blur level, using spectral envelope distribution features and importance information to generate a weight matrix, and employing neural networks to obtain analysis-based and model-based quality scores, allowing for differentiated quality assessment and processing of video frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single quality assessment method is used for all video frames, then the assessment process is simple, but the accuracy of quality evaluation deteriorates because fully-blurred and partially-blurred frames require different assessment approaches
Solution Approach 1:
The patent segments video frames into two categories: fully-blurred frames and partially-blurred frames. This segmentation allows the system to apply different quality assessment methods to different frame types, thereby improving measurement precision without requiring a completely complex unified system. The segmentation is based on detecting blur levels and classifying frames accordingly.
Solution Approach 2:
The patent implements a dynamic quality assessment system that adapts its methodology based on the input frame characteristics. The system dynamically selects between analysis-based assessment (for fully-blurred frames) and model-based assessment (for partially-blurred frames), making the assessment process flexible and context-dependent rather than static and one-size-fits-all.
2Reliability
If uniform quality assessment is applied to all frames, then processing is efficient, but reliability deteriorates because sub-area characteristics and blur levels are not considered
Solution Approach 1:
The patent applies local quality assessment by dividing frames into sub-areas and evaluating blur characteristics locally in each sub-area. This allows the system to consider spatial variations in blur levels and apply appropriate assessment methods to different regions, thereby improving reliability. The weight matrix generated from spectral envelope distribution reflects these local characteristics.
Solution Approach 2:
The patent performs partial quality assessment by focusing computational resources on relevant frame regions and characteristics. For fully-blurred frames, it uses analysis-based assessment that may not require full-frame detailed analysis, while for partially-blurred frames, it applies model-based assessment only to affected sub-areas, thereby maintaining reliability while improving processing efficiency.
3Measurement precision
If detailed spectral envelope analysis is performed for each sub-area, then quality assessment accuracy improves, but computational complexity and processing time increase
Solution Approach 1:
The patent segments the frame into multiple sub-areas and performs spectral envelope analysis on each segment independently. This segmentation allows parallel processing of different regions, reducing overall processing time while maintaining detailed analysis accuracy for each sub-area. The segmented approach also enables selective processing based on detected blur characteristics.
Solution Approach 2:
The patent performs preliminary classification of frames into fully-blurred or partially-blurred categories before conducting detailed spectral envelope analysis. This preliminary action allows the system to skip detailed sub-area analysis for frames that don't require it, or to focus detailed analysis only on relevant sub-areas, thereby reducing overall processing time while maintaining measurement precision where needed.
Data Source
AI summary
A video quality assessment apparatus and method are provided. The video quality assessment apparatus includes a memory storing one or more instructions; and a processor configured to execute the one or more instructions stored in the memory to: identify whether a frame included in a video is a fully-blurred frame or a partially-blurred frame based on a blur level of the frame, obtain, in response to the frame being the fully-blurred frame, an analysis-based quality score with respect to the fully-blurred frame; obtain, in response to the frame being the partially-blurred frame, a model-based quality score with respect to the partially-blurred frame; and process the video based on at least one of the analysis-based quality score or the model-based quality score to obtain a processed video.