Video Block Partitioning With Intra Prediction for Better Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for multimedia data, such as video, exceeds the bandwidth capabilities of existing channels, necessitating improved compression efficiency in image encoding and decoding, particularly in intra prediction mode information and block division methods.
Innovation Solution
A method and apparatus for encoding and decoding video signals that involve dividing blocks into partial blocks based on symmetrical or asymmetrical shapes, determining most probable modes, and adjusting intra prediction modes based on neighboring blocks, allowing for variable block sizes and shapes, and using multiple intra prediction modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If conventional encoding methods are used, then device complexity is low, but compression efficiency is insufficient to meet increasing multimedia data demands
Solution Approach 1:
The current block is divided into multiple partial blocks (first and second partial blocks) with different intra prediction modes applied to each segment. This segmentation allows for more precise local prediction, improving compression efficiency by better adapting to local image characteristics while maintaining manageable encoding complexity through systematic processing of each segment.
Solution Approach 2:
Different intra prediction modes (first intra prediction mode for the first partial block, second intra prediction mode for the second partial block) are applied to different regions of the current block based on their specific characteristics. This local quality approach improves compression efficiency by using the most appropriate prediction mode for each region, rather than applying a single mode uniformly across the entire block.
2Loss of energy
If multiple intra prediction modes are used for different partial blocks, then compression efficiency improves, but the amount of information to be encoded increases
Solution Approach 1:
The division information indicating how the current block is divided into partial blocks is encoded first, before encoding the intra prediction mode information for each partial block. This preliminary action allows the decoder to understand the block structure in advance, enabling more efficient subsequent decoding of prediction mode information and reducing overall information overhead.
Solution Approach 2:
Intra prediction mode information is selectively encoded for only certain partial blocks rather than all blocks. This partial action approach improves compression efficiency by focusing computational resources on blocks that benefit most from multiple prediction modes, while reducing information overhead by omitting redundant encoding for blocks where a single mode suffices.
3Loss of energy
If division information and multiple prediction modes are encoded, then image compression efficiency enhances, but transmission bandwidth requirements increase
Solution Approach 1:
Multiple intra prediction modes are applied selectively to certain partial blocks rather than uniformly to all blocks. This partial action strategy enhances compression efficiency for complex regions while minimizing additional data transmission volume by using simpler single-mode encoding for regions where it suffices, thus optimizing the trade-off between compression performance and bandwidth consumption.
Solution Approach 2:
The intra prediction mode parameters are dynamically changed based on the characteristics of each partial block. By adapting the prediction mode parameters to local image characteristics rather than using fixed parameters, the system achieves better compression efficiency that reduces overall data transmission volume, as the encoded data more accurately represents the original image content.
Data Source
AI summary
According to the present invention, a method of decoding a video signal may include decoding division information that indicates whether a current decoding block is divided into two partial blocks, decoding information that indicates a division direction for the current decoding block when the division information indicates that the current decoding block is divided into the two partial blocks, and dividing the current decoding block into the two partial blocks according to the division direction.