HEVC CABAC Lossless Coding with Bypass Mode for Higher Throughput
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High bit-rate encoding and decoding in HEVC using CABAC is computationally complex due to the use of multiple probability models, leading to significant processing resource consumption and time, especially in high throughput scenarios.
Innovation Solution
Implementing a high throughput binarization mode that uses a bypass coding mode with a simplified signing structure and adaptive binarization tables, and reducing the number of context models used in lossless coding, to bypass computationally expensive context estimation and probability update stages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple probability models are used in CABAC encoding, then encoding precision is improved, but processing complexity increases
Solution Approach 1:
The patent changes the parameter of probability model selection by introducing a flag that indicates whether to use context-adaptive probability models or fixed probability models. This allows dynamic adjustment of modeling complexity based on data characteristics, maintaining encoding precision when needed while reducing complexity when possible.
Solution Approach 2:
The patent implements dynamic switching between different coding modes (context-adaptive mode and bypass mode) based on data characteristics. The encoder can adaptively choose whether to update probability models for each symbol or use fixed models, making the system flexible in managing the trade-off between precision and complexity.
2Measurement precision
If context-adaptive probability models are updated for each symbol, then encoding accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies partial context-adaptive modeling by selectively updating probability models only for certain symbols or under certain conditions, rather than updating all models for every symbol. This reduces the total number of updates and processing time while maintaining sufficient encoding accuracy for the most important data elements.
Solution Approach 2:
The patent segments the encoding process into two distinct modes: context-adaptive mode for high-precision encoding of critical data, and bypass mode for faster encoding of less critical data. This segmentation allows the system to allocate processing resources efficiently, spending more time on important elements and less on others.
3Measurement precision
If context estimation and probability update stages are performed, then coding precision is improved, but throughput decreases
Solution Approach 1:
The patent implements dynamic mode switching that allows the encoder to transition between high-precision context-adaptive coding and high-speed bypass coding based on data characteristics and throughput requirements. This enables the system to maintain high throughput by using bypass mode for suitable data while still achieving good precision when context-adaptive coding is applied.
Solution Approach 2:
The patent changes the operational parameters of the encoder by introducing a controllable flag that switches between two distinct coding regimes. This parameter change allows flexible adjustment of the trade-off between precision and throughput, enabling the system to optimize for throughput when necessary while preserving precision capabilities when needed.
4Loss of information
If multiple context models are used in lossless coding, then compression ratio is improved, but device complexity increases
Solution Approach 1:
The patent changes the parameter of model selection by using a flag to indicate whether to apply context-adaptive probability models or fixed probability models for lossless coding. This allows the system to reduce model complexity and associated device requirements when using fixed models, while still achieving good compression ratios when context-adaptive models are deployed.
Data Source
AI summary
A system utilizing a high throughput lossless coding mode for CABAC in HEVC is described. The system may include an electronic device configured to obtain a block of data to be encoded using an arithmetic based encoder; determine whether the block of data is to be encoded using lossless encoding; in response to determining that the block of data is not to be encoded using lossless encoding, use a first Absolute-3 coding technique to encode the block of data; in response to determining that the block of data is to be encoded using lossless encoding, use a second Absolute-3 coding technique to encode the block of data; wherein the second Absolute-3 coding technique is different than the first Absolute-3 coding technique.