Video Block Splitting Decision Using Pre-assigned Quantization Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding methods using rate-distortion optimization (RDO) are not optimal when quantization parameters assigned to small blocks in a pre-encoding stage do not align with the larger blocks used during encoding, leading to suboptimal trade-offs between video quality and bitrate.
Innovation Solution
A method that computes rate-distortion costs for both splitting and non-splitting configurations of a current block, using previously assigned quantization parameters to determine whether the block should be split, allowing for efficient decision-making without re-performing the pre-encoding stage, and employing a Lagrangian parameter to impact the splitting decision based on coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If quantization parameters are assigned to small blocks in a pre-encoding stage, then the encoding speed is improved, but the rate-distortion optimization becomes suboptimal when larger blocks are used during encoding
Solution Approach 1:
The patent applies preliminary action by assigning quantization parameters to small blocks during a pre-encoding stage before the actual encoding process. This allows the encoder to quickly access pre-computed QP values during encoding, improving encoding speed while maintaining the ability to perform accurate rate-distortion optimization by considering multiple QP options for larger blocks.
Solution Approach 2:
The patent segments the picture into small blocks for pre-encoding analysis, assigning quantization parameters to each small block independently. This segmentation enables the encoder to later combine these small block QP assignments into larger blocks during rate-distortion optimization, allowing both fine-grained pre-processing and coarse-grained encoding decisions.
2Adaptability or versatility
If multiple quantization parameters are assigned to each large block from small block QP assignment, then the flexibility for rate-distortion optimization is improved, but the complexity of determining the best configuration increases
Solution Approach 1:
The patent applies local quality by allowing different quantization parameters to be assigned to different regions within a large block, based on the pre-computed small block QP assignments. This enables the encoder to select the most appropriate QP for each local region during rate-distortion optimization, improving adaptability while using systematic methods to manage the complexity of evaluating multiple configurations.
Solution Approach 2:
The patent introduces dynamics by making the quantization parameter assignment flexible and adaptive during the encoding process. Instead of using a fixed QP for each large block, the system dynamically selects from multiple possible QP configurations derived from small block assignments, allowing the encoder to adapt to local content characteristics while managing complexity through efficient search strategies.
3Productivity
If the encoder uses larger blocks for encoding, then the compression efficiency is improved, but the alignment with pre-assigned small block quantization parameters deteriorates
Solution Approach 1:
The patent merges the small block quantization parameter assignments with large block encoding by combining multiple small block QP values into candidate configurations for each large block. This merging process allows the encoder to maintain alignment with the detailed small block analysis while achieving the compression efficiency benefits of larger encoding blocks, resolving the misalignment issue between pre-encoding and encoding stages.
Data Source
AI summary
A method and an apparatus for encoding a picture are disclosed. Such method comprises: —determining a rate-distortion cost (41) for a current block of said picture when said current block is not split into subblocks, taking into account quantization parameters assigned to the subblocks of said current block, —determining (42) whether said current block is split or not according at least to the determined rate-distortion cost, —encoding (43) the current block according to the result of determining whether a current block is split or not into subblocks.