Binary Codeword Generation for HEVC Transform Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current High Efficiency Video Coding (HEVC) systems face challenges in efficiently encoding and decoding large input images and macroblocks due to high video data density, requiring improved methods for coding transform coefficients within transform units.
Innovation Solution
A method is provided for determining binary codewords for transform coefficients using a Truncated Rice code with a predefined maximum number of bits, where flags are coded for coefficients with absolute values greater than one or two, and symbols are converted into binary codewords based on a parameter variable that is updated after each conversion, optimizing the encoding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional binary codeword generation methods are used for transform coefficients, then encoding and decoding can be performed, but the complexity and memory requirements increase significantly
Solution Approach 1:
The patent segments the transform coefficient encoding process into distinct stages: significance flag coding for coefficients with absolute value greater than one, additional flag coding for coefficients with absolute value greater than two, and truncated rice code for remaining coefficients. This segmentation allows each stage to use optimized coding strategies, reducing overall complexity while maintaining encoding efficiency.
Solution Approach 2:
The patent introduces a parameter variable that is updated after each symbol conversion based on the syntax element of the most recently converted symbol. This dynamic parameter adjustment optimizes the truncated rice code generation process, reducing memory requirements and computational complexity while preserving the ability to handle high video data density effectively.
2Productivity
If high video data density is handled with traditional coding methods, then compression can be achieved, but processing speed decreases
Solution Approach 1:
The patent performs preliminary classification of transform coefficients by their absolute values before encoding. Coefficients are pre-categorized into groups (those with absolute value greater than one, greater than two, and others), allowing the encoding process to skip unnecessary calculations and directly apply the most efficient coding method to each group, thereby increasing processing speed without compromising compression efficiency.
Solution Approach 2:
The patent employs a dynamic parameter variable that adapts during the encoding process based on the syntax elements of previously processed symbols. This dynamic adjustment allows the system to optimize the truncated rice code generation in real-time, improving processing speed while maintaining the ability to handle high video data density effectively.
3Quantity of substance
If memory requirements for codeword storage are increased, then more coefficients can be handled, but the system becomes less efficient
Solution Approach 1:
The patent extracts and processes transform coefficients in a systematic sequence, handling them in groups based on their significance flags rather than storing all coefficients simultaneously in memory. By extracting and encoding coefficients in a structured manner (first those with absolute value greater than one, then greater than two, then the remainder), the system reduces memory requirements while maintaining the ability to handle large numbers of transform coefficients.
Data Source
AI summary
A system is provided for creating binary codewords for transform coefficients used for relating transform units (TUs) divided into coding units (CUs) in a High Efficiency Video Coding (HEVC) system. The system provides binarization of the codewords and removes unnecessary operations to reduce system complexity and increase compression performance. The system generates transform coefficients that relate the TUs and begins by providing a parameter variable (cRiceParam) set to an initial value of zero. Significant transform coefficients are converted into binary codewords based on the current value of the parameter variable, and the parameter variable is then updated with a new current value after each transform coefficient has been converted. Updating can be provided with reference to table values or the values can be provided from combination logic.