Video Quality Monitoring Using Bitstream Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The ITU-T G.1070 model, used for estimating perceptual quality of video, is unsuitable for monitoring as it does not accurately represent actual coding bit rates, frame rates, and packet loss rates, and fails to consider video content complexity, leading to inaccurate quality scores that do not correlate well with user subjective experience.
Innovation Solution
A system for perceptual quality monitoring that estimates coding bit rate, frame rate, and packet loss rate parameters from an encoded video bitstream, using a sliding window approach to collect data and normalize bit rates based on video frame complexity, providing accurate quality scores that align better with user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the G.1070 model uses assumed coding bit rate and frame rate parameters for quality estimation, then the model can operate as a planning tool, but the quality scores do not accurately reflect actual video quality and user experience
Solution Approach 1:
The system uses the encoded bitstream itself to extract actual coding bit rate and frame rate parameters through parsing SBV and VPS sequences, eliminating the need for external assumption inputs. The decoder automatically measures these parameters from the incoming stream, making the quality estimation self-serve based on real conditions rather than planned parameters.
Solution Approach 2:
The system continuously monitors actual coding bit rate and frame rate from the encoded bitstream and feeds this information back into the quality estimation model. This feedback loop ensures the model uses real-time actual parameters rather than static assumed values, continuously updating quality scores to reflect current video quality conditions.
2Reliability
If the G.1070 model uses assumed packet loss rate parameters, then the model can estimate expected quality, but it cannot accurately monitor actual packet loss and its impact on quality
Solution Approach 1:
The system extracts actual packet loss rate information directly from the RTP packets in the encoded bitstream by comparing expected versus actual packet arrival. This self-measurement approach eliminates the need for external packet loss monitoring systems, allowing the quality estimation model to directly incorporate real packet loss data for accurate quality assessment.
3Measurement precision
If the G.1070 model does not consider video content complexity, then the model remains simple to implement, but it produces inaccurate quality scores for scenes with varying complexity
Solution Approach 1:
The system uses decoder feedback signals including complexity indicators from the encoded bitstream to adjust quality estimation. The decoder provides information about actual video content complexity through parsed sequences, allowing the model to dynamically adapt quality scores based on the specific content being decoded rather than using generic assumptions.
Solution Approach 2:
The system changes the quality estimation parameters dynamically based on actual video content characteristics. By adjusting the weightings and calculations in the quality model according to measured content complexity from the bitstream, the system adapts to different video scenarios (simple vs. complex scenes) to produce more accurate quality scores that correlate with subjective user experience.
4Measurement precision
If the system extracts actual bitstream parameters for quality estimation, then quality monitoring accuracy improves, but the processing time and computational load increase
Solution Approach 1:
The system performs preliminary extraction of coding bit rate and frame rate parameters during the normal decoding process by parsing SBV and VPS sequences. These parameters are prepared in advance as part of the standard decoding workflow, so no additional separate measurement step is required, and the quality estimation can proceed using already-available data from the decoding process.
Solution Approach 2:
The system merges the quality parameter extraction operations with the existing video decoding process. By combining the parsing of coding parameters, frame rate measurement, and quality estimation into a unified processing pipeline, the system eliminates separate processing stages and reduces overall computational overhead while maintaining accurate quality monitoring.
Data Source
AI summary
Systems and methods of perceptual quality monitoring of video information, communications, and entertainment that can estimate the perceptual quality of video with high accuracy, and can be used to produce quality scores that better correlate with subjective quality scores of an end user. The systems and methods of perceptual quality monitoring of video can generate, from an encoded input video bitstream, estimates of one or more quality parameters relating to the video, such as the coding bit rate parameter, the video frame rate parameter, and the packet loss rate parameter, and provide these video quality parameter estimates to a predetermined video quality estimation model. Because the estimates of the video quality parameters are generated from the encoded input video bitstream as it is being received, the systems and methods are suitable for use as QoE monitoring tools.