Picture Coding Sub-Block Context Derivation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current picture coding and decoding technologies face challenges in efficiently processing differential coefficient information due to dependency between context index calculation and decoding, leading to high processing delays and calculation complexities, especially in real-time applications.

Innovation Solution

The proposed solution involves partitioning differential information into sub-blocks, encoding significant sub-block information, differential coefficient information, and deriving contexts based on neighboring sub-block information, allowing for parallel processing and reduced calculation complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If context index calculation and decoding are performed sequentially to ensure accurate probability estimation, then coding efficiency is improved, but processing delay increases

Engineering Contradiction:
Improvecoding efficiencyVSAvoidprocessing delay
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent segments the differential coefficient block into multiple sub-blocks and processes them independently in parallel. Each sub-block has its own context index calculation and decoding operations, eliminating sequential dependencies while maintaining accurate probability estimation through localized context modeling.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary calculation of context indexes for all sub-blocks before actual decoding begins. By pre-computing context indexes using neighboring block information that is already available, the system eliminates the need for sequential waiting during decoding, thereby reducing processing delay while maintaining coding efficiency.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If context index calculation depends on decoded results to achieve high probability estimation accuracy, then coding precision is improved, but device complexity increases

Engineering Contradiction:
Improveprobability estimation accuracyVSAvoidcalculation complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies local quality by using neighboring sub-block information specific to each sub-block's location to calculate context indexes. Each sub-block uses locally relevant decoded results from its neighbors rather than global decoded results, maintaining high probability estimation accuracy while reducing calculation complexity through localized processing.

Inventive Principle:
Principle #3Local quality

3Measurement precision

If all significant differential coefficient information is processed sequentially for accurate decoding, then decoding accuracy is improved, but processing speed decreases

Engineering Contradiction:
Improvedecoding accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The patent divides the differential coefficient processing into multiple independent sub-block segments. Each sub-block can be decoded in parallel using its pre-calculated context index, maintaining decoding accuracy through localized context modeling while dramatically improving processing speed through parallel execution.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary calculation of context indexes for all sub-blocks before decoding begins. This pre-computation enables all sub-blocks to be decoded simultaneously without waiting for sequential context index calculations, thereby improving processing speed while maintaining decoding accuracy through accurate pre-computed context probabilities.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS9712839B2Picture coding device, picture coding method, and picture coding program, and picture decoding device, picture decoding method, and picture decoding program
Publication Date: 2017.07.18 JVC KENWOOD CORP
  • US9712839B2 patent drawing
  • US9712839B2 patent drawing
  • US9712839B2 patent drawing

AI summary

A picture coding device partitions differential information between a coding target picture and a prediction target picture into sub blocks and codes the partitioned sub blocks. The device codes significant sub block information that represents whether or not all the values of differential coefficients belonging to the sub block are zero and codes significant differential coefficient information that represents whether or not the value of the differential coefficient is zero. A context index processing unit derives a context used for coding the significant sub block information and a context used for coding the significant differential coefficient information of the differential coefficient that is a coding target based on an addition equation using the significant sub block information of a coded sub block that is horizontally neighboring to the coding target sub block and the significant sub block information of a coded sub block that is vertically neighboring.