CABAC Significance Map Scanning for Parallel Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression techniques, such as H.264 and HEVC, face challenges in parallel processing due to dependencies between elements in the same scanning line during context model selection for significance maps, which limits their efficiency in encoding and decoding video data.
Innovation Solution
The proposed solution involves choosing context models for elements in a significance map based on neighboring elements from other scanning lines, eliminating dependencies within the same scanning line, and using pre-determined or adaptable scanning patterns to facilitate parallel processing in video encoding and decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If context models are selected based on neighboring elements in the same scanning line, then coding precision is improved through better adaptation, but processing speed deteriorates due to sequential dependencies
Solution Approach 1:
The patent divides the significance map into multiple scanning lines and processes them independently in parallel. By selecting context models based on neighboring elements from other scanning lines rather than the same scanning line, the patent creates independent processing segments that eliminate sequential dependencies within each scanning line, enabling parallel processing while maintaining adaptive context model selection.
2Productivity
If adaptive context model selection is implemented, then compression efficiency is improved, but device complexity increases
Solution Approach 1:
The patent applies different context model selection strategies to different scanning lines based on their specific characteristics. By adapting context models locally to each scanning line's neighboring elements from other lines, the patent achieves better compression efficiency while managing complexity through localized rather than universal adaptation.
Solution Approach 2:
The patent pre-determines scanning patterns and context model selection rules before processing the significance map. This preliminary preparation of scanning patterns and context model configurations reduces the computational complexity during actual encoding/decoding operations while maintaining adaptive compression efficiency.
3Ease of manufacture
If sequential processing of scanning lines is used, then implementation simplicity is maintained, but productivity decreases
Solution Approach 1:
The patent segments the significance map processing into independent scanning line units that can be processed in parallel. Each scanning line is treated as an independent processing unit with its own context model selection based on neighboring elements from other lines, enabling parallel execution while maintaining relatively simple implementation of each individual unit.
Solution Approach 2:
The patent transitions from one-dimensional sequential processing of scanning lines to two-dimensional parallel processing by utilizing spatial relationships across different scanning lines. By selecting context models from neighboring elements in other scanning lines, the patent introduces a dimensional approach that enables parallel processing without significantly increasing implementation complexity.
Data Source
AI summary
In various embodiments, a significance map of a matrix of video data coefficients is encoded or decoded using context-based adaptive binary arithmetic coding (CABAC). The significance map scanned line-by-line along a scanning pattern. Each line may be a vertical, horizontal, or diagonal section of the scanning pattern. Context models for each element processed in a particular line are chosen based on values of neighboring elements that are not in the line. The neighboring elements may be limited to those contained within one or two other scanning lines. Avoiding reliance on neighbors that are in the same scanning line facilitates parallel processing.