Intra Prediction Mode Segmentation for Video Decoding Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards, such as ITU-T H.264 and HEVC, face limitations in efficiently coding a large number of intra prediction modes, particularly in handling non-directional and directional prediction modes, which affects coding efficiency and compression capabilities.
Innovation Solution
The techniques described involve methods for encoding and decoding intra prediction data by determining whether the selected intra prediction mode corresponds to a most probable mode candidate, and if not, categorizing it into sets of modes other than the most probable candidates, along with filtering and transform processes to optimize intra prediction coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a large number of intra prediction modes are supported to improve prediction accuracy, then coding efficiency is improved, but the complexity of decoding the prediction mode increases
Solution Approach 1:
The patent segments the intra prediction modes into two distinct categories: most probable mode candidates (MPMs) and non-MPM modes. This segmentation allows the decoder to handle different mode types with different decoding strategies, reducing overall complexity while maintaining support for a large number of modes. The segmentation is implemented through the `intra_mode_flag` and `intra_mode_idx` syntax elements that differentiate between MPM and non-MPM cases.
Solution Approach 2:
The patent extracts the most probable mode candidates from the full set of intra prediction modes and handles them separately through a dedicated decoding path. By taking out the MPM cases, the decoder can use more efficient decoding mechanisms for these frequent modes while using a different, more comprehensive approach for less frequent non-MPM modes, thus reducing overall decoding complexity.
2Productivity
If more intra prediction modes are added to improve compression capability, then coding efficiency is improved, but the bitstream complexity increases
Solution Approach 1:
The patent applies local quality by providing different coding representations for different mode types within the same bitstream. MPM modes use a compact representation through `intra_mode_idx` directly, while non-MPM modes use a different representation through `intra_mode_flag` and `intra_mode_idx`. This localized differentiation optimizes the bitstream for each mode type's frequency and complexity characteristics.
Solution Approach 2:
The patent changes the parameter representation strategy based on mode probability. For MPM modes, the system uses direct index representation, while for non-MPM modes, it uses flag-based representation. This parameter change approach allows efficient encoding of a large number of modes without uniformly increasing bitstream complexity for all modes.
3Ease of manufacture
If directional and non-directional prediction modes are handled uniformly, then implementation simplicity is maintained, but coding efficiency decreases
Solution Approach 1:
The patent applies local quality by providing different coding representations for different mode types within the same bitstream. MPM modes use a compact representation through `intra_mode_idx` directly, while non-MPM modes use a different representation through `intra_mode_flag` and `intra_mode_idx`. This localized differentiation optimizes the bitstream for each mode type's frequency and complexity characteristics.
Solution Approach 2:
The patent introduces dynamic adaptability in the decoding process through the `intra_mode_flag` mechanism. The decoder dynamically adjusts its behavior based on the flag value: when true, it decodes MPM modes with one set of operations; when false, it decodes non-MPM modes with different operations. This dynamic approach enables the system to adapt to different mode types without requiring separate implementations for each.
Data Source
AI summary
A video coding device may be configured to perform intra prediction coding according to one or more of the techniques described herein.