Image Coding Mixing Reference Picture Information

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The conventional method for coding video data using reference picture lists in video coding standards like H.264/AVC results in decreased coding efficiency due to the increased amount of information required for identifying reference pictures.

Innovation Solution

An image coding method that mixes first and second information to generate a mixed signal, selects a variable-length coding table based on the maximum variation range of this signal, and codes the mixed signal using the selected table, reducing the coding amount of the information identifying reference picture lists and pictures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If first information and second information are used separately to identify reference picture lists and reference pictures, then the reference picture identification is complete and accurate, but the coding efficiency decreases due to the increased amount of information

Engineering Contradiction:

Improvecoding efficiencyVSAvoidamount of information

Core Design Contradiction:

Loss of informationVSQuantity of substance

Solution Approach 1:

The patent combines first information (reference picture list identification) and second information (reference picture identification) into a single mixed signal. This merging reduces the total amount of information to be coded while maintaining the complete reference picture identification capability, thereby improving coding efficiency without losing essential information.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The mixed signal serves multiple functions simultaneously: it identifies both the reference picture list and the reference picture within that list. This multi-functionality eliminates the need for separate coding of first and second information, reducing the overall information quantity while maintaining complete reference picture identification.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Ease of manufacture

If separate variable-length coding tables are used for first information and second information, then the coding is straightforward and maintains good compatibility, but the coding efficiency is reduced due to the increased information amount

Engineering Contradiction:

Improvecoding implementation simplicityVSAvoidcoding efficiency

Core Design Contradiction:

Ease of manufactureVSLoss of information

Solution Approach 1:

The patent merges the coding of first information and second information into a single variable-length coding process using one mixed signal. This approach maintains implementation simplicity by using a single coding table and process, while simultaneously improving coding efficiency by reducing the total information amount compared to separate coding approaches.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS9497458B2Image coding method, image decoding method, image coding apparatus, image decoding apparatus, program, and integrated ciruit

Publication Date: 2016.11.15 SUN PATENT TRUST

AI summary

An image coding method includes: obtaining first information and second information from a current signal to be coded, the first information identifying at least one reference picture list used in inter prediction of an image out of a plurality of reference picture lists, the second information identifying at least one reference picture used in the inter prediction of the image out of reference pictures included in the at least one reference picture list identified by the first information; mixing the first information and the second information to generate a mixed signal; selecting a variable-length coding table according to a maximum variation range of the mixed signal; and variable-length coding the mixed signal using the selected variable-length coding table.