Block-Based Prediction Using Partition Shape and Mode Derivation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding/decoding technologies struggle to efficiently split blocks based on their size and shape for improved prediction, leading to suboptimal prediction efficiency and accuracy.
Innovation Solution
The proposed method involves splitting a target block into multiple partition blocks, determining a prediction mode for each partition block based on its size and shape, and performing prediction accordingly, using Most Probable Mode (MPM) lists and reconstructed pixels as reference samples.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a block is split into multiple partition blocks based on size and shape, then prediction efficiency and accuracy are improved, but device complexity and processing overhead increase
Solution Approach 1:
The patent applies segmentation by dividing a target block into multiple partition blocks based on size and shape characteristics. The encoder determines optimal partitioning patterns (e.g., symmetric asymmetric, asymmetric symmetric, asymmetric asymmetric) and applies different prediction modes to each partition block, thereby improving prediction accuracy while managing complexity through structured segmentation rules
Solution Approach 2:
The patent implements local quality by deriving prediction modes specifically for each partition block based on its local characteristics. Different prediction modes (intra prediction, inter prediction, merge mode, skip mode) are selected for different partition blocks according to their size, shape, and content properties, allowing localized optimization of prediction accuracy
2Productivity
If different prediction modes are derived for each partition block, then prediction efficiency is enhanced, but encoding complexity and processing time increase
Solution Approach 1:
The patent applies preliminary action by pre-defining a set of candidate prediction modes and partitioning patterns before actual encoding. The encoder prepares Most Probable Mode (MPM) lists and candidate partition patterns in advance, then selects from these pre-prepared options during encoding, reducing real-time computation while maintaining efficiency
Solution Approach 2:
The patent implements parameter changes by dynamically adjusting prediction mode parameters and partitioning parameters based on block characteristics. The encoder changes parameters such as partition size, prediction mode type, and MPM list composition according to the specific size and shape of each target block, optimizing the balance between efficiency and encoding time
Data Source
AI summary
Disclosed herein are a video decoding method and apparatus and a video encoding method and apparatus. In video encoding and decoding, multiple partition blocks are generated by splitting a target block. A prediction mode is derived for at least a part of the multiple partition blocks, among the multiple partition blocks, and prediction is performed on the multiple partition blocks based on the derived prediction mode. When prediction is performed on the partition blocks, information related to the target block may be used, and information related to an additional partition block, which is predicted prior to the partition block, may be used.