Prediction Direction Selection in HEVC Image Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The High Efficiency Video Coding (HEVC) protocol is excessively complex, leading to high requirements on encoding apparatus performance, which limits its application in achieving real-time image encoding due to inaccurate prediction direction selection and high computational complexity.
Innovation Solution
A prediction direction selection method that determines a reference direction based on texture consistency between coding units, using preliminarily selected direction vectors and adjusting candidate direction vectors to ensure minimal cost prediction directions, thereby improving accuracy and reducing computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If the HEVC coding protocol is used for image encoding, then image compression efficiency is improved, but encoding complexity increases excessively
Solution Approach 1:
The patent segments the prediction direction selection process into two distinct phases: preliminary selection that filters out obviously suboptimal directions, and accurate selection that precisely determines the optimal direction from the reduced candidate set. This segmentation reduces the overall computational complexity while maintaining compression efficiency.
Solution Approach 2:
The patent performs preliminary selection of prediction directions before the accurate selection phase. By pre-filtering candidate directions based on coarse criteria (such as gradient directions or neighboring block predictions), the system reduces the search space for the subsequent accurate selection, thereby lowering encoding complexity while preserving compression performance.
2Measurement precision
If 35 prediction directions are traversed to ensure accurate prediction direction selection, then prediction accuracy is improved, but encoding speed decreases
Solution Approach 1:
The patent divides the prediction direction evaluation into two stages: a preliminary stage that quickly eliminates obviously poor directions using simple criteria, and an accurate stage that performs detailed evaluation only on the remaining candidate directions. This segmentation maintains prediction accuracy while significantly improving encoding speed.
Solution Approach 2:
Instead of evaluating all 35 prediction directions equally, the patent performs partial action by focusing computational resources only on the most promising candidate directions identified in the preliminary selection phase. This approach achieves sufficient prediction accuracy without the excessive computational cost of evaluating all possible directions.
3Measurement precision
If reconstruction is performed for prediction direction selection, then prediction accuracy is improved, but computational complexity and execution time increase
Solution Approach 1:
The patent performs preliminary selection of prediction directions using simplified criteria that do not require full reconstruction. By using proxy metrics such as gradient analysis or neighboring block predictions, the system identifies candidate directions without the time-consuming reconstruction process, thereby reducing execution time while maintaining the ability to achieve accurate predictions in the subsequent selection phase.
Data Source
AI summary
A prediction direction selection method includes constructing, according to texture consistency of a prediction unit relative to an upper-layer coding unit of the coding unit in which the prediction unit is located, a reference direction corresponding to the prediction unit. The method also includes obtaining, for the coding unit, preliminarily selected direction vectors that are selected from intra-frame directions according to rate-distortion costs. The method next includes determining processed candidate direction vectors by using the reference direction and the preliminarily selected direction vectors. The processed candidate direction vectors are used for selecting a direction corresponding to an optimal cost as a prediction direction for unit prediction performed for the coding unit.