Parallel Entropy Coding with Context-Based Bitstream Partitioning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current entropy coding methods, such as H.264's CAVLC and CABAC, face challenges in efficiently handling situations where the number of parallel decoding processing units differs from the number of distinct estimated probabilities, leading to computational demands and limited processing power issues, especially in high-quality video decoding.

Innovation Solution

A method and architecture that utilize parallel entropy coders to encode an input sequence into multiple subsequences, which are then combined into a single bitstream, allowing for parallel decoding and interleaving based on a context model to reconstruct the original sequence, enabling efficient encoding and decoding even with varying processing unit availability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of substance

If CABAC is used for entropy coding to achieve greater compression, then compression efficiency is improved, but computational complexity increases

Engineering Contradiction:
Improvecompression efficiencyVSAvoidcomputational complexity
Core Design Contradiction:
Loss of substanceVSDevice complexity

Solution Approach 1:

The patent segments the entropy coding process into multiple parallel contexts, where each context handles a specific subset of probability models. This allows the complex CABAC algorithm to be divided into manageable parallel tasks that can be executed simultaneously, reducing the computational burden on single processing units while maintaining compression efficiency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces parallelism as a new dimension for processing entropy-coded data. By organizing multiple entropy coding contexts to operate in parallel rather than sequentially, the system achieves both improved compression efficiency and reduced computational complexity through distributed processing across multiple units.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Productivity

If the number of parallel decoding processing units is increased to improve decoding speed, then processing throughput is improved, but device complexity and resource requirements increase

Engineering Contradiction:
Improvedecoding speedVSAvoidprocessing unit configuration
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent designs a universal parallel processing architecture where multiple decoding processing units can handle different entropy coding contexts using the same fundamental decoding logic. This multi-functional design allows the system to scale decoding speed by adding processing units without proportionally increasing device complexity, as each unit performs similar functions with different parameter configurations.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If the number of parallel decoding processing units differs from the number of distinct estimated probabilities, then flexibility in resource allocation is improved, but mapping complexity increases

Engineering Contradiction:
Improveresource allocation flexibilityVSAvoidmapping complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces a context model as an intermediary layer between the parallel decoding processing units and the probability estimates. This context model acts as a mediator that maps multiple probability estimates to a manageable number of processing units, or vice versa, thereby enabling flexible resource allocation while reducing the direct mapping complexity between processing units and probability models.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2362657B1Parallel entropy coding and decoding methods and devices
Publication Date: 2013.04.24 BLACKBERRY LTD
  • EP2362657B1 patent drawingFigure 1
  • EP2362657B1 patent drawingFigure 2
  • EP2362657B1 patent drawingFigure 3

AI summary

A method for entropy coding data using parallel entropy coders to encode an input sequence as a plurality of encoded subsequences, which are then combined to form an output bitstream. The input sequence is separated into subsequences on the basis of a context model. A method for entropy decoding a bitstream of encoded data by extracting a plurality of encoded subsequences from the bitstream and entropy decoding the encoded subsequences in parallel to generate a plurality of decoded subsequences, which are then interleaved based on a context model to produce a reconstructed sequence.