Motion Vector Candidate Filtering for Temporal Matching in Image Coding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional image coding methods fail to appropriately code images when blocks are referenced from pictures that temporally match, leading to inefficiencies in coding efficiency.

Innovation Solution

An image coding method that determines whether to add a motion vector to a list based on temporal matching between pictures, selecting a motion vector predictor and coding the difference between the current block's motion vector and the predictor, while excluding vectors from temporally matching pictures to avoid unnecessary scaling and improve coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If motion vectors from all blocks are added to the list regardless of temporal matching, then the number of motion vector candidates increases, but coding efficiency deteriorates due to unnecessary scaling operations

Engineering Contradiction:
Improvenumber of motion vector candidatesVSAvoidcoding efficiency
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent extracts and removes motion vectors from blocks where the associated picture and reference picture temporally match from the candidate list. This prevents unnecessary motion vector scaling operations while maintaining adequate candidate diversity for accurate motion prediction.

Inventive Principle:
Principle #2Taking out (Extraction)

2Adaptability or versatility

If motion vectors from temporally matching pictures are included in the list, then more candidates are available for prediction, but unnecessary scaling operations increase processing complexity

Engineering Contradiction:
Improveprediction candidate availabilityVSAvoidprocessing complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

Motion vectors from blocks with temporal matching between associated and reference pictures are extracted and excluded from the candidate list, eliminating unnecessary scaling operations while preserving sufficient prediction candidates.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameter of temporal matching detection to identify and exclude cases where associated picture and reference picture temporally match, thereby modifying the composition of the motion vector candidate list to reduce processing complexity.

Inventive Principle:
Principle #35Parameter changes

3Ease of manufacture

If all motion vectors are processed without temporal matching checks, then the coding process is simpler, but image coding accuracy deteriorates

Engineering Contradiction:
Improvecoding process simplicityVSAvoidimage coding accuracy
Core Design Contradiction:
Ease of manufactureVSManufacturing precision

Solution Approach 1:

The patent performs preliminary temporal matching checks between associated pictures and reference pictures before adding motion vectors to the candidate list. This preliminary action identifies and excludes vectors that would require unnecessary scaling, improving coding accuracy while maintaining process simplicity.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS9210402B2Image coding method, image decoding method, image coding apparatus, image decoding apparatus, and image coding and decoding apparatus
Publication Date: 2015.12.08 SUN PATENT TRUST
  • US9210402B2 patent drawing
  • US9210402B2 patent drawing
  • US9210402B2 patent drawing

AI summary

An image coding method includes: (i) determining, for each of one or more associated blocks, whether to add a motion vector of the associated block to a list, and (ii) adding the motion vector of the associated block to the list when determining that the motion vector of the associated block is to be added to the list; selecting a motion vector from the list as a motion vector predictor; and (i) coding a current block using a motion vector of the current block, and (ii) coding a difference between the motion vector of the current block and the motion vector predictor, and in the determining, it is determined that the motion vector of the associated block is not to be added to the list when an associated picture and an associated reference picture match temporally or when a current picture and a current reference picture match temporally.