Adaptive Loop Filter Coefficient Encoding for Lower Video Coding Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing loop filters, such as the adaptive loop filter (ALF), in video coding standards like VVC and HEVC, suffer from high complexity and time-consuming calculations, which affect their practical application.
Innovation Solution
Implement a loop filtering method that determines multiple sets of filter coefficients using non-differential encoding and decoding methods, and employs non-exponential Golomb encoding for clip index parameters, optimizing the encoding and decoding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the adaptive loop filter (ALF) is used to improve filtering effect and coding efficiency, then the filtering quality is improved, but the calculation complexity and time consumption increase significantly
Solution Approach 1:
The patent changes the parameter representation method by using non-differential encoding instead of differential encoding for filter coefficients, and non-exponential Golomb encoding instead of exponential Golomb encoding for clip index parameters. This parameter encoding transformation reduces calculation complexity while maintaining filtering quality.
Solution Approach 2:
The patent replaces the complex differential encoding and exponential Golomb encoding methods with simpler non-differential encoding and non-exponential Golomb encoding methods, which are computationally cheaper and faster to execute, achieving the same filtering function with reduced complexity.
2Productivity
If the adaptive loop filter (ALF) is used to improve filtering effect, then the coding efficiency is improved, but the time consumption increases significantly
Solution Approach 1:
The patent transforms the encoding parameters from differential encoding to non-differential encoding, and from exponential Golomb encoding to non-exponential Golomb encoding. This parameter transformation maintains coding efficiency while significantly reducing the time consumption in encoding and decoding operations.
Solution Approach 2:
The patent substitutes the computationally intensive differential encoding mechanism with a simpler non-differential encoding mechanism, replacing complex calculation operations with more efficient encoding operations that achieve the same functional result with less time consumption.
3Measurement precision
If differential encoding and exponential Golomb encoding are used for filter coefficients and clip index parameters, then the encoding precision is maintained, but the calculation complexity increases
Solution Approach 1:
The patent replaces expensive differential encoding and exponential Golomb encoding methods with cheaper non-differential encoding and non-exponential Golomb encoding methods. These simpler encoding methods maintain the necessary precision for filter coefficients and clip index parameters while reducing encoding complexity and computational overhead.
Solution Approach 2:
The patent changes the encoding parameters from differential encoding to non-differential encoding, and from exponential Golomb encoding to non-exponential Golomb encoding. This parameter change preserves encoding precision for video reconstruction while significantly simplifying the encoding process and reducing computational complexity.
Data Source
Figure 1~2
Figure 3~4
Figure 5A~5B
AI summary
A loop filtering method and device are provided. By optimizing the encoding method in the process of encoding and decoding loop filtering, the calculation complexity of loop filtering is reduced, the calculation speed is accelerated, and the encoding and decoding performance is improved. A loop filtering method includes determining multiple sets of filter coefficients for a loop filter, and encoding the multiple sets of filter coefficients by using a non-differential encoding method. Another loop filtering method includes determining a clip index parameter of a loop filter, and encoding the clip index parameter by using a non-exponential Golomb encoding method. Still another loop filtering method includes determining a loop filter clip parameter of a brightness component and a loop filter clip parameter of a chrominance component of an image frame, where the loop filter clip parameter of the brightness component and the loop filter clip parameter of the chrominance component are parameters selected from the same parameter list.