Two-Stage Intra Prediction Mode Search for Image Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding and decoding technologies face challenges in efficiently determining intra prediction modes, leading to suboptimal compression and reconstruction of images.
Innovation Solution
An image decoding method and apparatus that determine a first intra prediction mode from a bitstream, establish a search range for a second intra prediction mode, and generate a prediction block using the second mode to improve image reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a full search of all intra prediction modes is performed to find the optimal mode, then prediction accuracy is improved, but computational complexity and encoding time increase
Solution Approach 1:
The patent applies preliminary action by performing a first intra prediction using a default or previously determined mode before conducting the second refined prediction. This preliminary prediction provides an initial reference that guides the subsequent search, reducing the need to evaluate all possible modes exhaustively while maintaining prediction accuracy.
Solution Approach 2:
The patent segments the intra prediction process into two distinct stages: a first intra prediction phase that establishes a baseline, and a second intra prediction phase that refines the result within a constrained search range. This segmentation allows each phase to focus on specific optimization goals, improving overall efficiency without sacrificing accuracy.
2Productivity
If a constrained search range is used for the second intra prediction mode, then encoding speed is improved, but prediction accuracy may deteriorate
Solution Approach 1:
The first intra prediction serves as a preliminary action that establishes a reference point. The search range for the second intra prediction is determined based on the result of this first prediction, ensuring that the constrained search focuses on the most promising modes while maintaining accuracy.
Solution Approach 2:
The search range for the second intra prediction mode is dynamically determined based on the first intra prediction result. This dynamic adjustment allows the encoding process to adapt to the specific characteristics of each block, searching more aggressively in directions where improvement is most likely while maintaining speed.
3Manufacturing precision
If multiple intra prediction modes are evaluated for each block, then image quality is improved, but processing time increases
Solution Approach 1:
The evaluation of intra prediction modes is segmented into two phases: an initial evaluation using a first prediction mode, followed by a refined evaluation using a second prediction mode selected from a constrained search range. This segmentation reduces the total number of modes that require full evaluation while maintaining image quality.
Solution Approach 2:
The first intra prediction acts as a preliminary filtering step that identifies promising prediction directions. This preliminary action reduces the search space for the second prediction, allowing multiple modes to be evaluated only when necessary and significantly reducing overall processing time.
Data Source
AI summary
An image decoding method and apparatus, and an image encoding method and apparatus, are provided. The image decoding method includes obtaining information about a first intra prediction mode of a current block from a bitstream, determining the first intra prediction mode of the current block based on the information about the first intra prediction mode, determining a search range of a second intra prediction mode based on based on the first intra prediction mode, determining the second intra prediction mode for the current block within the search range of the second intra prediction mode, and generating a prediction block by performing intra prediction on the current block by using the second intra prediction mode, and generating a reconstructed image based on the prediction block.