Palette Mode Run Entropy Coding With Adaptive Golomb Order

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing HEVC standard's Palette mode encoding for video sequences faces inefficiencies in encoding the 'Run' syntax element, particularly due to the fixed Golomb order used for entropy coding, which does not adapt well to varying prediction modes and block characteristics.

Innovation Solution

Adapting the Golomb order based on the prediction mode and optionally other parameters, such as the palette size and coding unit size, to optimize the entropy coding of the 'Run' syntax element in the Palette mode, allowing for more efficient encoding by adjusting the expected range of Run values.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a fixed Golomb order is used for entropy coding of the Run syntax element, then the encoding process is simple, but the encoding efficiency is poor due to inability to adapt to varying prediction modes and block characteristics

Engineering Contradiction:
Improveencoding efficiencyVSAvoidcoding complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies dynamics by making the Golomb order variable rather than fixed. The Golomb order is adapted based on the prediction mode (first prediction mode vs. second prediction mode) and can be further adjusted based on block characteristics such as block size or palette size. This dynamic adaptation allows the encoding system to optimize for different scenarios, improving encoding efficiency while maintaining reasonable complexity through conditional selection.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter (Golomb order) based on different prediction modes and block characteristics. When the first prediction mode is used, a first Golomb order is applied; when the second prediction mode is used, a second Golomb order is applied. This parameter change strategy enables the system to adapt to varying statistical properties of the Run syntax element under different coding conditions, resolving the contradiction between simplicity and efficiency.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If the Golomb order is adapted based on prediction mode and block characteristics, then encoding efficiency is improved, but the complexity of determining the appropriate Golomb order increases

Engineering Contradiction:
Improvecompression performanceVSAvoidparameter determination complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements parameter changes by selecting different Golmob orders based on the prediction mode and block characteristics. The system determines whether to use a first or second prediction mode, and accordingly applies a first or second Golomb order. This can be further extended to adjust based on block size or palette size, creating a hierarchical parameter selection process that balances complexity and performance.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies local quality by using different Golomb orders for different local conditions (different prediction modes or different block characteristics). Instead of using a single global Golomb order, the system tailors the Golomb order to the specific local context of each block being encoded, whether that context is defined by prediction mode, block size, or palette size. This localized adaptation improves compression performance without requiring complex global optimization.

Inventive Principle:
Principle #3Local quality

3Productivity

If a fixed encoding approach is used for all prediction modes, then the coding process is straightforward, but the bitrate is higher due to suboptimal compression

Engineering Contradiction:
Improvebitrate efficiencyVSAvoidcoding process complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent changes the encoding parameter (Golomb order) based on the prediction mode to improve bitrate efficiency. By applying a first Golomb order for the first prediction mode and a second Golomb order for the second prediction mode, the system optimizes the compression for each mode's specific characteristics, reducing the overall bitrate compared to a fixed encoding approach.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces dynamics into the encoding process by adapting the Golomb order to the prediction mode. This dynamic adjustment allows the system to respond to different coding scenarios, selecting the most appropriate entropy coding parameter for each prediction mode, thereby improving bitrate efficiency while maintaining a relatively simple decision-based structure.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3080988B1Parameter derivation for entropy coding of a syntax element
Publication Date: 2022.04.27 CANON KK
  • EP3080988B1 patent drawingFigure 1
  • EP3080988B1 patent drawingFigure 2
  • EP3080988B1 patent drawingFigure 3

AI summary

The present invention concerns a method and a device for predicting a displacement vector component syntax element encoding in the process of encoding or decoding a video. It particularly concerns the "Run" syntax element in a palette mode encoding in HEVC. It applies more particularly to a mode of coding where a block of pixel is predictively encoded based on a predictor block encoded with, or built from, a so-called palette. It proposes to improve the encoding of the Run syntax element by using a variable Golomb order. Several embodiments are describes where the Golomb order is adapted according to different parameters. Accordingly, the coding of the Run syntax element is improved.