Scalable Video Encoding With Adaptive Color Component Bit Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression technologies fail to adapt to picture content having different color characteristics.
Innovation Solution
The proposed solution is to divide the video signal into a first signal component and a second signal component, and a third signal component, and perform entropy encoding on these components to adapt to the picture content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed proportions of bit rates are allocated to Y, U, and V components, then the encoding process is simple, but large distortions occur in encoded pictures due to inability to adapt to different color characteristics
Solution Approach 1:
The patent applies dynamics by making the bit rate allocation proportional to the variance of each color component rather than using fixed proportions. The system dynamically adjusts the bit rate allocation based on the actual variance characteristics of Y, U, and V components in different pictures, allowing the encoding process to adapt to varying color characteristics while maintaining simplicity through automated variance-based calculation
Solution Approach 2:
The patent changes the parameter of bit rate allocation proportion from fixed values to variable values determined by component variance. By calculating the variance of each color component and allocating bit rates proportionally to these variance values, the system achieves adaptive bit rate distribution that matches the actual data characteristics of different pictures
2Productivity
If deep learning is applied to end-to-end picture encoding and decoding, then compression efficiency is improved, but the system complexity increases significantly
Solution Approach 1:
The patent segments the deep learning model into separate modules: a feature extraction network that processes input pictures to generate feature maps, and a separate bit rate allocation mechanism that operates on the variance of color components. This segmentation allows the complex deep learning components to be modularized, improving compression efficiency while managing system complexity through functional separation
Solution Approach 2:
The patent introduces feature maps as an intermediary between the input picture and the bit rate allocation process. The feature extraction network generates these intermediate feature representations, which then serve as the basis for calculating component variance and determining optimal bit rate allocation. This intermediary layer simplifies the overall system by providing a structured intermediate representation that bridges complex processing and control functions
Data Source
AI summary
This application provides a scalable encoding and decoding method and apparatus. The encoding method includes: applying a control signal of a first signal component of a video signal to a first feature map of the first signal component, to obtain a second feature map of the first signal component, where the control signal of the first signal component is obtained through learning; applying a control signal of a second signal component of the video signal to a first feature map of the second signal component, to obtain a second feature map of the second signal component, where the control signal of the second signal component is obtained through learning; and obtaining a bitstream of the video signal based on the second feature map of the first signal component and the second feature map of the second signal component.