HEVC Intra Prediction Quantization Offset Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In High Efficiency Video Coding (HEVC), inter predicted and intra predicted coding units within the same largest coding unit must share the same quantization parameter due to delta quantization being signaled at the LCU level, leading to inefficiencies as intra predicted units require a lower quantization parameter, resulting in an unnecessary increase in the size of the LCU.
Innovation Solution
Introducing a quantization offset parameter to allow intra predicted coding units to have a lower quantization parameter than inter predicted units within the same LCU, without explicitly signaling a delta quantization parameter change, by modifying the quantization parameter based on the prediction mode type.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If intra predicted and inter predicted coding units within the same LCU share the same quantization parameter, then the signaling structure is simplified, but the coding efficiency deteriorates due to unnecessary increase in LCU size
Solution Approach 1:
The patent applies local quality by allowing different quantization parameters for intra predicted and inter predicted coding units within the same LCU. Specifically, the quantization parameter for intra predicted CUs is derived by adding a quantization offset parameter to the base quantization parameter, while inter predicted CUs use the base quantization parameter directly. This enables localized optimization of compression for different prediction types without requiring complex per-CU signaling.
Solution Approach 2:
The patent changes the quantization parameter based on prediction mode type by introducing a quantization offset parameter. The encoder selectively applies this offset to intra predicted coding units while leaving inter predicted coding units unchanged. This parameter differentiation improves coding efficiency by allowing tighter quantization for intra predicted blocks which generally require higher quality, while maintaining coarser quantization for inter predicted blocks.
2Reliability
If the quantization parameter for inter predicted coding units is lowered to match intra predicted coding units, then the subjective quality of inter predicted regions is improved, but the overall image data size increases unnecessarily
Solution Approach 1:
The patent applies local quality by applying different quantization strategies to different regions based on prediction type. Intra predicted coding units receive a lower quantization parameter (base QP + offset) for higher quality, while inter predicted coding units use the base quantization parameter for smaller size. This localized differentiation ensures that quality improvements are applied only where necessary rather than uniformly across the entire image.
Solution Approach 2:
The patent changes the quantization parameter selectively based on prediction mode. The quantization offset parameter is added to the base quantization parameter only for intra predicted coding units, creating a parameter differentiation that optimizes the trade-off between quality and data size for each prediction type without uniformly increasing overall image size.
3Adaptability or versatility
If a delta quantization parameter is explicitly signaled at the LCU level, then the quantization parameter can be adjusted for different prediction types, but the overhead of signaling increases
Solution Approach 1:
The patent achieves local quality adjustment through a quantization offset parameter that is applied selectively to intra predicted coding units. This offset enables different quantization behavior for different prediction types without requiring separate delta quantization parameters for each coding unit, thereby reducing signaling overhead while maintaining the ability to optimize quality locally.
Solution Approach 2:
The quantization offset parameter serves as a universal mechanism that can be applied to multiple intra predicted coding units within an LCU simultaneously. This multi-functional approach allows the system to adjust quantization for all intra predicted blocks using a single parameter value, avoiding the need for individual delta quantization signaling for each block and thereby reducing overall signaling overhead.
Data Source
AI summary
A method of encoding an image into a coded image, the method comprising: writing a quantization offset parameter into the coded image, determining a prediction mode type for coding a block of image samples of the image into a coding unit of the coded image, determining a quantization parameter for the block of image samples, and determining if the prediction mode type is of a predetermined type, wherein if the prediction mode type is of the predetermined type, the method further comprises: modifying the determined quantization parameter using the quantization offset parameter, and performing a quantization process for the block of image samples using the modified quantization parameter, and wherein if the prediction mode type is not of the predetermined type, the method further comprises: performing a quantization process for the block of image samples using the determined quantization parameter.