Video Block Prediction Mode Pruning for Boundary Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding technologies face inefficiencies due to longer code lengths and lower encoding efficiency when using Semi Fixed-Length Code for residual values, particularly at block boundaries where reference pixels are unavailable, leading to encoding redundancy.
Innovation Solution
The proposed methods involve removing unavailable prediction modes from the initial mode list, filling reference pixels using available neighbors, and optimizing boundary sign bits in fixed-length codes to improve encoding and decoding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If Semi Fixed-Length Code is used to encode residual values, then code length is reduced, but encoding efficiency remains lower and code length is still longer
Solution Approach 1:
The patent changes the encoding parameters by switching from Semi Fixed-Length Code to Fixed-Length Code for residual values, and optimizes the prediction mode selection process. This parameter change achieves both shorter code length and improved encoding efficiency by eliminating the inefficiencies of the previous encoding scheme
Solution Approach 2:
The patent extracts and removes the prediction mode information from the bitstream when reference pixels are unavailable, as this information would be redundant. By taking out unnecessary encoding elements, the overall code length is reduced while maintaining encoding efficiency
2Adaptability or versatility
If all prediction modes are kept in the initial prediction mode list, then prediction flexibility is maintained, but encoding redundancy increases
Solution Approach 1:
The patent applies local quality by adjusting the prediction mode list based on the specific location and availability of reference pixels. When reference pixels are unavailable, the prediction mode list is locally optimized by removing modes that cannot be executed, reducing encoding redundancy while maintaining flexibility for available modes
Solution Approach 2:
The patent extracts and removes unavailable prediction modes from the initial prediction mode list when reference pixels are not available. This eliminates encoding redundancy for modes that cannot be performed, while retaining the necessary prediction flexibility through the remaining valid modes
3Adaptability or versatility
If reference pixels outside the first boundary are not available, then boundary prediction is limited, but filling reference pixels from the second boundary increases prediction ability
Solution Approach 1:
The patent uses reference pixels from the second boundary as an intermediary source to fill the missing reference pixels outside the first boundary. This mediator approach enables boundary prediction by transferring available pixel data across boundaries, increasing prediction ability without requiring direct access to unavailable reference pixels
Solution Approach 2:
The patent performs preliminary action by filling reference pixels outside the first boundary with data from the second boundary before the actual prediction process. This pre-filling ensures that all necessary reference data is available upfront, enabling complete prediction functionality without increasing operational complexity during decoding
Data Source
AI summary
The present invention provides an encoding and decoding method, apparatus and device. The method comprises: when a reference pixel of a current block is not available, removing a target prediction mode from an initial prediction mode list corresponding to the current block, so as to obtain a target prediction mode list; and encoding and decoding the current block on the basis of the target prediction mode list. The present invention can improve the encoding performance and the decoding performance.