Fractional-Pel Block Vectors for Higher-Efficiency Video Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques, such as HEVC and VVC, limit coding gains for non-screen content by using block vectors with integer-pel resolution, which are inadequate for natural content.
Innovation Solution
Implementing block vectors with fractional-pel resolution to identify positions between samples in the same picture, allowing for improved prediction accuracy and coding efficiency in Intra Block Copy (IBC) and Intra Template Matching (IntraTMP) modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If block vectors with integer-pel resolution are used, then device complexity is reduced and ease of operation is improved, but coding efficiency deteriorates for natural content
Solution Approach 1:
The patent changes the resolution parameter of block vectors from integer-pel to fractional-pel (e.g., half-pel, quarter-pel), allowing the system to adapt to different content types. This parameter change enables finer positioning between samples, improving prediction accuracy for natural content while maintaining compatibility with existing video coding frameworks
2Productivity
If block vectors with fractional-pel resolution are used, then coding efficiency is improved for natural content, but device complexity increases
Solution Approach 1:
The patent introduces dynamic block vector resolution, where the system can switch between integer-pel and fractional-pel modes depending on the content type. For screen content, integer-pel mode is used to maintain simplicity, while for natural content, fractional-pel mode is activated to improve coding efficiency, making the system adaptive rather than static
Solution Approach 2:
The patent segments the video content processing into different modes based on content type detection. The system divides the coding process into screen content regions (using integer-pel block vectors) and natural content regions (using fractional-pel block vectors), allowing each segment to be processed with the most appropriate method
3Ease of operation
If integer-pel block vectors are used, then ease of operation is maintained, but measurement precision deteriorates for positions between samples
Solution Approach 1:
The patent changes the precision parameter of block vectors by introducing fractional-pel resolution (e.g., half-pel, quarter-pel, eighth-pel). This allows the system to identify positions between integer sample points, effectively increasing measurement precision from integer coordinates to fractional coordinates, similar to changing from meters to millimeters for more precise positioning
Data Source
AI summary
A method of encoding or decoding video data includes determining a block vector for a current block of a current picture, the block vector identifying a fractional-pel position between reference samples in the current picture and having a fractional-pel resolution that is less than integer-pel resolution; generating a prediction block from reference samples in the current picture based on the fractional-pel position; and encoding or decoding the current block based on the prediction block.