Aspect-Ratio Adaptive Intra-Prediction for Rectangular Video Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video coding standards, such as HEVC/H.265, fail to efficiently handle rectangular video coding blocks due to the asymmetry in the number of directional intra-prediction modes, which does not account for the aspect ratio or availability of reference samples, leading to increased computational complexity and reduced coding efficiency.
Innovation Solution
An apparatus and method that extend the set of directional intra-prediction modes based on the aspect ratio of rectangular video coding blocks, enabling or disabling modes based on reference sample availability, and signaling these modes through mode mapping and a one-bit flag, thereby improving prediction accuracy and reducing complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the existing set of predetermined directional intra-prediction modes is used for rectangular video coding blocks, then the coding implementation is simple, but the coding efficiency is reduced due to asymmetry not accounting for aspect ratio
Solution Approach 1:
The patent applies dynamics by making the set of directional intra-prediction modes adaptive rather than static. The apparatus dynamically selects and extends prediction modes based on the aspect ratio of rectangular video coding blocks, allowing the system to adjust its behavior according to block geometry. This resolves the contradiction by maintaining simple implementation through standardized procedures while improving coding efficiency through context-adaptive mode selection.
Solution Approach 2:
The patent applies local quality by tailoring the intra-prediction mode set to specific local characteristics of video blocks. Different aspect ratios (vertical, horizontal, square) receive different extended mode sets optimized for their geometric properties. This allows each local region (block type) to use the most appropriate prediction modes, improving overall coding efficiency without complicating the global implementation structure.
2Device complexity
If the existing set of predetermined directional intra-prediction modes is used, then the computational complexity is reduced, but the prediction accuracy is lowered
Solution Approach 1:
The patent applies parameter changes by modifying the set of available prediction mode parameters based on block aspect ratio. For vertical rectangular blocks, additional modes with angles optimized for vertical geometry are enabled; for horizontal blocks, modes suited for horizontal orientation are added. This parameter adaptation improves prediction accuracy for each block type while maintaining manageable computational complexity through systematic mode extension rather than exhaustive search.
3Productivity
If the intra-prediction modes are extended based on aspect ratio, then the coding efficiency is improved, but the device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the aspect ratio determination into discrete categories (vertical rectangular, horizontal rectangular, square blocks). Each category has a predefined extended mode set, avoiding the need for continuous complex calculations. This segmentation approach improves coding efficiency through targeted mode extension while limiting apparatus complexity by using classification-based rather than continuous optimization approaches.
4Ease of operation
If directional intra-prediction modes are selected without considering reference sample availability, then the mode selection is simple, but the prediction reliability is reduced
Solution Approach 1:
The patent applies preliminary action by determining reference sample availability before finalizing intra-prediction mode selection. The apparatus preliminarily identifies which reference samples are available based on block position and geometry, then uses this information to guide mode extension and selection. This preliminary assessment improves prediction reliability by ensuring selected modes have adequate reference data, while maintaining ease of operation through systematic pre-evaluation rather than complex real-time decisions.
Data Source
AI summary
An intra-prediction apparatus and method are provided for extending a set of predetermined directional intra-prediction modes of a rectangular video coding block. The intra-prediction apparatus determines an aspect ratio of the rectangular video coding block, and determines a complementary angle range based on the aspect ratio. The apparatus then selects a complementary direction within the complementary directional range, and adds an extension directional intra-prediction mode associated with the complementary direction to the set of predetermined directional intra-prediction modes. Based on the extension directional intra-prediction mode, the apparatus intra-predicts pixel values of pixels of the rectangular video coding block.