Motion-Based Adaptive Quantization for Video Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image compression technologies face challenges in reconciling the opposing requirements of maintaining high quality for static content and reducing bandwidth for dynamic content in video streaming, particularly in 'desktop streaming' scenarios where still portions require sharper contrasts and moving content needs less visual detail.
Innovation Solution
The implementation of motion-based adaptive quantization, where the quantization parameter (QP) for image data is adjusted from a start QP to a target QP in a multi-step change, specifically increasing compression quality for static regions and decreasing it for non-static regions, allowing for customized compression based on the content's motion status.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a single quantization parameter is used for all regions, then device complexity is reduced, but manufacturing precision of compression quality cannot be achieved for different content types
Solution Approach 1:
The patent applies local quality by using different quantization parameters for different regions of the video content. Specifically, static regions (desktop content) use one QP value while dynamic regions (video content) use another QP value, allowing each region to be compressed with the appropriate quality level for its content type.
Solution Approach 2:
The patent implements dynamics by making the quantization parameter adaptive rather than fixed. The system dynamically selects different QP values based on the detected motion characteristics of each region, transitioning between static and dynamic content types to apply the most appropriate compression settings.
2Manufacturing precision
If high compression quality is applied to all regions, then manufacturing precision of image quality is improved, but loss of energy in transmission bandwidth increases
Solution Approach 1:
The patent applies local quality by using different quantization parameters for different regions of the video content. Specifically, static regions (desktop content) use one QP value while dynamic regions (video content) use another QP value, allowing each region to be compressed with the appropriate quality level for its content type.
Solution Approach 2:
The patent implements parameter changes by adjusting the quantization parameter based on the content type. The system changes the QP value from a first value for static regions to a second value for dynamic regions, optimizing the balance between quality and bandwidth consumption for each region.
3Loss of energy
If low compression quality is applied to all regions, then loss of energy in transmission bandwidth is reduced, but manufacturing precision of static content quality deteriorates
Solution Approach 1:
The patent applies local quality by using different quantization parameters for different regions of the video content. Specifically, static regions (desktop content) use one QP value while dynamic regions (video content) use another QP value, allowing each region to be compressed with the appropriate quality level for its content type.
4Manufacturing precision
If quantization parameter is increased for static regions, then manufacturing precision of compression quality is improved, but loss of substance in bit rate increases
Solution Approach 1:
The patent applies local quality by using different quantization parameters for different regions of the video content. Specifically, static regions (desktop content) use one QP value while dynamic regions (video content) use another QP value, allowing each region to be compressed with the appropriate quality level for its content type.
Solution Approach 2:
The patent implements parameter changes by adjusting the quantization parameter based on the content type. The system changes the QP value from a first value for static regions to a second value for dynamic regions, optimizing the balance between quality and bandwidth consumption for each region.
Data Source
AI summary
A method and apparatus for compressing a data stream comprising a plurality of pictures are described. A first quantization parameter (QP) from a plurality of QPs is determined, for a static region in a current picture. The plurality of QPs change in accordance with a multi-step change from a start QP to a target QP and each one of the plurality of QPs is to be applied to a respective one from successive static regions in successive pictures. In response to determining, based upon statistics on static region(s) in the current picture which are associated with the first QP, that the first QP is selected, compression of the static region is caused based upon the first QP, and in response to determining that the first QP is not selected, compression of the static region is caused based upon a second QP that is greater than the first QP.