Entropy Decoding Context Variable Storage for Dependent Slice Segments

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video codecs face challenges in efficiently encoding and decoding high-resolution video content, particularly in managing context variables across slice segments during entropy decoding, which affects data size and compression efficiency.

Innovation Solution

The proposed method utilizes context information from nearby data to optimize entropy decoding, storing context variables after the last maximum coding unit of each slice segment is decoded, allowing for efficient synchronization and reduced data size by obtaining initial variables from previously decoded independent slice segments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If context variables are stored for all slice segments, then decoding accuracy is improved, but data size increases

Engineering Contradiction:
Improvedecoding accuracyVSAvoiddata size
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent applies local quality by differentiating the treatment of context variables based on slice segment type. Independent slice segments initialize context variables to default values, while dependent slice segments reuse context variables from previous segments. This selective approach optimizes decoding accuracy only where needed (independent segments) while minimizing data size (dependent segments).

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent segments slice segments into two categories: independent and dependent. This segmentation allows the system to apply different context variable management strategies to each type, storing context variables only for independent segments while reusing them for dependent segments, thus balancing decoding accuracy with data size reduction.

Inventive Principle:
Principle #1Segmentation

2Quantity of substance

If context variables are reused across slice segments, then data size is reduced, but synchronization complexity increases

Engineering Contradiction:
Improvedata sizeVSAvoidsynchronization complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-defining the conditions under which context variables are stored (independent slice segments) and reused (dependent slice segments). The decoder is预先 configured to recognize slice segment types and apply the appropriate context variable management strategy, simplifying the synchronization process despite the complexity of reuse.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If context variables are initialized for each slice segment, then decoding independence is improved, but processing time increases

Engineering Contradiction:
Improvedecoding independenceVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies local quality by applying initialization only where it provides value (independent slice segments) and skipping it where it would be redundant (dependent slice segments). This selective initialization maintains decoding independence for independent segments while reducing processing time for dependent segments.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP2940999B1Method for entropy-decoding slice segments
Publication Date: 2019.06.12 SAMSUNG ELECTRONICS CO LTD
  • EP2940999B1 patent drawingFigure 1A~1B
  • EP2940999B1 patent drawingFigure 2A~2B
  • EP2940999B1 patent drawingFigure 3~4

AI summary

Provided are entropy encoding and entropy decoding for video encoding and decoding. The video entropy decoding method includes: determining a bin string and a bin index for a maximum coding unit that is obtained from a bitstream; determining a value of a syntax element by comparing the determined bin string with bin strings that is assignable to the syntax element in the bin index; storing context variables for the maximum coding unit when the syntax element is a last syntax element in the maximum coding unit, a dependent slice segment is includable in a picture in which the maximum coding unit is included, and the maximum coding unit is a last maximum coding unit in a slice segment; and restoring symbols of the maximum coding unit by using the determined value of the syntax element.