Adaptive Intra Prediction Mode Selection for Image Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image encoding and decoding technologies face challenges in efficiently compressing high-resolution and high-quality images, leading to increased data volumes and transmission/storage costs.
Innovation Solution
The method involves decoding information indicating whether to use a maximum of N intra prediction modes or a maximum of M intra prediction modes for a current block, deriving the intra prediction mode, and performing inverse-mapping when M modes are used, to generate an intra prediction block.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional image encoding and decoding technologies are used, then image data can be transmitted and stored, but the data volume increases significantly for high-resolution and high-quality images, leading to increased transmission and storage costs
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the maximum number of intra prediction modes (switching between M and N modes) based on block characteristics such as size, shape, and gradient. This adaptive parameter adjustment optimizes compression efficiency for different image regions, reducing overall data volume while maintaining quality, thereby lowering transmission and storage costs for high-resolution images.
2Measurement precision
If a maximum of N intra prediction modes are used for intra prediction, then prediction accuracy improves, but the complexity of encoding and decoding increases
Solution Approach 1:
The patent implements dynamics by making the maximum number of intra prediction modes adaptive rather than fixed. The encoder and decoder dynamically switch between using M modes and N modes based on block characteristics, allowing the system to adjust prediction accuracy and complexity in real-time according to the specific image content and block properties.
Solution Approach 2:
The patent applies local quality by differentiating the number of prediction modes used for different blocks based on their specific characteristics. Rather than using a uniform number of modes across all blocks, the system selectively applies M or N modes to different blocks depending on their size, shape, and gradient properties, optimizing both accuracy and complexity locally.
3Device complexity
If a maximum of M intra prediction modes are used for intra prediction, then encoding and decoding complexity is reduced, but prediction accuracy decreases
Solution Approach 1:
The patent uses dynamics to enable flexible switching between M modes and N modes based on block characteristics. This allows the system to reduce complexity to M modes when appropriate while maintaining the option to use N modes when higher accuracy is needed, creating a dynamic balance between complexity and prediction accuracy.
Solution Approach 2:
The patent applies parameter changes by adjusting the maximum number of intra prediction modes as a variable parameter rather than a fixed value. The system changes this parameter adaptively based on block size, shape, and gradient, allowing optimization of the balance between complexity and accuracy for different image regions.
Data Source
AI summary
An image encoding/decoding method and apparatus are provided. An image decoding method of the present invention comprises decoding first information indicating whether a maximum of N intra prediction modes are used or a maximum of M intra prediction modes are used for an intra prediction for a current block, in which the M is smaller than the N, deriving an intra prediction mode of the current block, inverse-mapping the derived intra prediction mode on a corresponding intra prediction mode among the maximum N intra prediction modes when the decoded first information indicates that the maximum M intra prediction modes are used for the intra prediction for the current block, and generating an intra prediction block by performing an intra prediction for the current block, based on the inverse-mapped intra prediction mode.