Video Encoder Prediction Modes With Hybrid RDO Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding methods face challenges in achieving both speed and accuracy due to the high resource and time requirements of full RDO algorithms, while fast RDO methods are less accurate when using original pixels without reconstructed pixels.
Innovation Solution
A video encoder and encoding method that performs first and second optimization operations using original and reconstructed pixels to generate prediction information, improving prediction mode determination through a combination of fast and full RDO circuits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full RDO algorithm is used for mode prediction, then prediction accuracy is improved, but hardware resource consumption and computation time increase significantly
Solution Approach 1:
The mode prediction process is divided into two stages: a fast RDO stage that evaluates candidate modes using simplified rate-distortion cost calculation (excluding coefficient syntax cost) to generate intermediate prediction modes, and a full RDO stage that performs complete evaluation only on selected modes. This segmentation allows the system to achieve both speed (through fast RDO filtering) and accuracy (through full RDO verification on promising candidates).
Solution Approach 2:
The fast RDO algorithm performs preliminary evaluation of multiple candidate modes before applying the more computationally intensive full RDO algorithm. By pre-filtering candidates based on rate-distortion cost using simplified metrics, the system prepares a reduced set of promising modes for final evaluation, thereby reducing overall computation time while maintaining prediction accuracy.
2Productivity
If fast RDO algorithm is used to save hardware resources and time, then encoding speed is improved, but prediction accuracy deteriorates when using original pixels instead of reconstructed pixels
Solution Approach 1:
The patent introduces reconstructed pixels from adjacent blocks as an intermediary element that bridges fast RDO and full RDO processes. The prediction circuit uses reconstructed pixels from adjacent blocks to enhance the accuracy of fast RDO calculations, creating an intermediate representation that combines the speed of fast RDO with improved accuracy through reconstructed pixel information.
Solution Approach 2:
The system dynamically changes the parameters used in rate-distortion cost calculation based on the processing stage and available data. When using fast RDO, the system uses simplified parameters excluding coefficient syntax cost. When full RDO is applied, complete parameters including coefficient syntax cost are used. This parameter adaptation allows the system to optimize for speed during fast RDO while ensuring accuracy during full RDO evaluation.
Data Source
AI summary
A video coding method includes the following steps: performing a first optimization operation on a sub-coding block to select original pixels or reconstructed pixels of an adjacent block of the sub-coding block according to a base prediction mode to generate an intermediate prediction mode of the sub-coding block; performing a second optimization operation on the sub-coding block to determine a prediction mode according to the intermediate prediction mode, the original pixels of the sub-coding block, and the reconstructed pixels of the adjacent block; generating prediction information of the sub-coding block according to the original pixels of the sub-coding block, the reconstructed pixels of the adjacent block, and the prediction mode; generating encoding coefficients and reconstructed pixels of the sub-coding block according to the prediction information of the sub-coding block; and generating a bit stream according to the encoding coefficients and the prediction information.