Prediction Mode Determination for Video Coding Pipeline Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face inefficiencies when CTUs are further split, particularly when child nodes have areas less than a threshold, leading to reduced pipeline processing speed and coding efficiency due to the use of different prediction modes for these nodes.
Innovation Solution
A prediction mode determining method and apparatus that determines a prediction mode status for child nodes based on their area and split manner, using intra or inter prediction modes, ensuring a consistent prediction mode across all child nodes to facilitate pipeline processing and improve coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If different prediction modes are used for child nodes with area less than threshold, then prediction accuracy is improved, but pipeline processing speed is reduced
Solution Approach 1:
The patent applies local quality by allowing different prediction mode handling for different regions: child nodes with area less than threshold use one approach (parsing prediction mode flag from bitstream), while child nodes with area greater than or equal to threshold use another approach (inheriting prediction mode from parent node). This regional differentiation resolves the contradiction by optimizing for accuracy where needed while maintaining speed where possible.
2Manufacturing precision
If different prediction modes are used for child nodes, then coding precision is improved, but device complexity increases
Solution Approach 1:
The patent implements local quality by applying different processing rules to different child nodes based on their area relative to a threshold. Child nodes below the threshold undergo complex processing (parsing prediction mode flags), while those at or above the threshold use simpler inheritance. This reduces overall device complexity while maintaining coding precision where it matters most.
Solution Approach 2:
The patent segments the processing logic into two distinct paths: one for small child nodes (area < threshold) that requires full processing, and another for larger child nodes (area >= threshold) that uses inheritance. This segmentation simplifies the overall system complexity by dividing the problem into manageable cases with different processing requirements.
3Measurement precision
If prediction mode is determined by parsing bitstream for each child node, then prediction accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by having child nodes with area greater than or equal to threshold inherit the prediction mode from their parent node without parsing the bitstream. This pre-determination of prediction mode for certain nodes eliminates unnecessary processing time while maintaining accuracy requirements for smaller nodes that do parse the bitstream.
Solution Approach 2:
The patent uses local quality by applying different time-costly operations to different nodes: full bitstream parsing for small child nodes where accuracy is critical, and fast inheritance for larger child nodes where time efficiency is prioritized. This resolves the time-accuracy tradeoff by localizing the expensive operation only where necessary.
Data Source
AI summary
A prediction mode determining method and apparatus are described. A prediction mode status of a child node of a target node is determined based on an area and a split manner of the target node, where the prediction mode status includes intra prediction mode only and inter prediction mode only. Further, a prediction mode of the child node is determined based on the prediction mode status of the child node. In this way, a same prediction mode is used for all child nodes obtained by splitting the target node. This facilitates pipeline processing of hardware, increases a subsequent processing speed, and improves video coding efficiency.