Video Coding Motion Candidate List Construction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing moving picture coding technologies face inefficiencies in coding motion information due to limited neighboring block references and increased processing load when comparing motion information, leading to suboptimal coding efficiency and increased coding amounts.

Innovation Solution

A moving picture coding device and method that constructs a list of motion information candidates by deriving temporal and spatial candidates, determines whether to use a merging prediction mode, and codes an index to select the appropriate motion information candidate, thereby reducing processing load and improving coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If motion information of multiple neighboring blocks is compared in detail to improve coding accuracy, then coding efficiency is improved, but processing load increases

Engineering Contradiction:

Improvecoding accuracyVSAvoidprocessing load

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the necessary motion information from neighboring blocks (spatial and temporal candidates) rather than comparing all possible motion information in detail. By selecting specific candidate blocks and extracting their motion vectors, the system achieves adequate coding accuracy while significantly reducing processing load.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements partial action by limiting the number of neighboring blocks that are compared and the depth of motion information analysis. Instead of exhaustive comparison, the system evaluates a predetermined number of spatial and temporal candidates, achieving acceptable coding efficiency with reduced processing requirements.

Inventive Principle:

Principle #16Partial or excessive action

2Measurement precision

If motion information is coded for each processing target block to ensure accuracy, then coding precision is maintained, but coding amount increases

Engineering Contradiction:

Improvecoding precisionVSAvoidcoding amount

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent uses motion information from neighboring blocks (spatial candidates A0, A1, B0, B1 and temporal candidate C) as copies or references for the current block's motion information. By copying motion vectors from these candidate blocks and using differential coding, the system maintains coding precision while significantly reducing the amount of data that needs to be transmitted.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent merges motion information from multiple sources (spatial neighboring blocks and temporal reference blocks) into a unified candidate list. This merging allows the system to select the best motion information while reducing redundant coding, as the same motion vectors can serve multiple blocks when appropriate.

Inventive Principle:

Principle #5Merging (Combining)

3Productivity

If limited neighboring blocks are used for prediction to reduce processing load, then processing efficiency is improved, but coding efficiency deteriorates

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidcoding efficiency

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent extends the prediction approach by considering motion information from two dimensions: spatial neighboring blocks (A0, A1, B0, B1) and temporal reference blocks (C). This multi-dimensional approach to gathering motion candidates improves coding efficiency without significantly increasing processing load, as the additional temporal dimension provides valuable motion information.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent creates a universal candidate list that can serve multiple prediction purposes. The same set of spatial and temporal candidates can be used for different block types and prediction modes, making the processing efficient while maintaining coding accuracy through the versatility of the candidate selection approach.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10171830B1Moving picture coding device, moving picture coding method, and moving picture coding program, and moving picture decoding device, moving picture decoding method, and moving picture decoding program

Publication Date: 2019.01.01 NEC CORP

AI summary

A merging motion information calculating unit calculates motion information of a plurality of coded neighboring blocks located at predetermined positions neighboring to a coding target block in space as spatial motion information candidates of the coding target block, in a case where there are spatial motion information candidates having the same motion information out of the spatial motion information candidates, sets one of the spatial motion information candidates having the same motion information as the spatial motion information candidate and, calculates a temporal motion information candidate of the coding target block by using the motion information of a coded block included in a picture that is different in time from a picture including the coding target block, and includes the spatial motion information candidates and the temporal motion information candidate in candidates for the motion information.