Video Quality Assessment Using Packet Header Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video quality assessment methods for network video services are inefficient due to high computational complexity, especially in real-time monitoring on terminal devices with low capabilities, and lack accuracy as they rely on parsing video bit streams or only consider packet loss rates.
Innovation Solution
A video quality assessment model that calculates quality using packet header information, eliminating the need for parsing video bit streams and incorporating features specific to video data, such as compression distortion and frame impairment, to provide a more accurate assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video quality assessment is performed by parsing video bit streams completely, then assessment accuracy is improved, but computational complexity increases making real-time assessment infeasible
Solution Approach 1:
The patent extracts only the necessary packet header information from video packets for quality assessment, rather than parsing the entire video bit stream. This extraction approach obtains key parameters like packet loss rates and encoding information from headers alone, achieving acceptable assessment accuracy while dramatically reducing computational complexity to enable real-time processing on terminal devices.
Solution Approach 2:
The patent performs partial parsing by selectively extracting only the most critical information from packet headers (such as loss rates and encoding parameters) rather than completely parsing all video data. This partial action approach provides sufficient quality assessment information while avoiding the excessive computational burden of full stream parsing.
2Device complexity
If packet loss rate alone is used for video quality assessment, then computational complexity is reduced, but assessment accuracy deteriorates
Solution Approach 1:
The patent merges multiple assessment dimensions by combining packet loss rate analysis with encoding parameter analysis (such as quantization parameters and frame types) extracted from packet headers. This combination approach maintains low computational complexity while improving assessment accuracy beyond what packet loss rate alone can provide, by considering both transmission quality and encoding quality factors.
Data Source
AI summary
A video data quality assessment method and apparatus are disclosed. The video data quality assessment method includes: acquiring a compression distortion parameter of video data; acquiring a frame impairment distortion parameter/video data rebuffering influence parameter; calculating a video quality parameter According to the compression distortion parameter and the frame impairment distortion parameter/video data rebuffering influence parameter, where the video quality parameter is a difference between the compression distortion parameter and the frame impairment distortion parameter/video data rebuffering influence parameter.