Motion Vector Prediction Using Neighboring Reference Lists

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video coding systems face challenges in reducing the bitrate associated with motion vectors, particularly in low-bitrate applications, as they require significant bandwidth for transmitting motion vector information, and existing Motion Vector Prediction (MVP) techniques may not adequately enhance the availability of motion vector predictors without additional side information.

Innovation Solution

A priority-based MVP scheme is developed that derives motion vector predictors or candidates from spatially neighboring blocks using motion vectors from both list 0 and list 1 reference pictures, allowing for enhanced predictor availability without the need for additional side information transmission, by considering different reference pictures and using pre-defined or adaptive priority orders for candidate selection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If motion vectors are transmitted for temporal prediction, then prediction accuracy is improved, but bitrate consumption increases significantly

Engineering Contradiction:
Improveprediction accuracyVSAvoidbitrate
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent uses motion vectors from spatially neighboring blocks as predictors for the current block's motion vector. Instead of transmitting complete motion vector information for each block, the system copies predictor information from neighboring blocks and only transmits the difference (residue), significantly reducing bitrate while maintaining prediction accuracy

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent introduces motion vector predictors as an intermediary element between neighboring blocks and the current block. These predictors serve as intermediate values that reduce the amount of information needed for accurate motion compensation, acting as a mediator that bridges spatial redundancy exploitation

Inventive Principle:
Principle #24Intermediary (Mediator)

2Quantity of substance

If Motion Vector Prediction (MVP) techniques are used to reduce bitrate, then bandwidth consumption decreases, but the availability of motion vector predictors is insufficient without additional side information

Engineering Contradiction:
ImprovebitrateVSAvoidpredictor availability
Core Design Contradiction:
Quantity of substanceVSAdaptability or versatility

Solution Approach 1:

The patent enables the decoder to derive motion vector predictors autonomously using only decoded information from neighboring blocks. The system is self-sufficient in generating predictors without requiring additional side information transmission, as the decoder can independently reconstruct the same predictor set as the encoder using the established priority-based selection rules

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent segments the motion vector prediction process into distinct priority levels and candidate categories. By dividing the predictor selection into structured segments with predefined priorities, the system ensures consistent predictor availability across different coding scenarios without requiring additional signaling

Inventive Principle:
Principle #1Segmentation

3Device complexity

If existing MVP techniques consider only MV with the same reference picture list and index, then implementation complexity is reduced, but the availability of spatial MVP candidates is limited

Engineering Contradiction:
Improveimplementation complexityVSAvoidcandidate availability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent introduces dynamic adaptability into the MVP candidate selection process. The system dynamically adjusts which neighboring blocks can serve as MVP candidates based on their reference picture list and index compatibility with the current block. This dynamic selection mechanism increases candidate availability while maintaining manageable implementation complexity through clear selection rules

Inventive Principle:
Principle #15Dynamics

4Productivity

If motion vector residues are reduced to improve coding efficiency, then compression performance increases, but more sophisticated prediction methods are required

Engineering Contradiction:
Improvecoding efficiencyVSAvoidprediction method complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent changes the parameters of motion vector prediction by introducing priority-based candidate selection and considering multiple reference picture lists. These parameter changes enable more accurate prediction and smaller residues, improving coding efficiency. The complexity increase is mitigated by systematic parameter organization and clear selection hierarchies

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2643970B1Method and apparatus of spatial motion vector prediction
Publication Date: 2018.04.11 HFI INNOVATION INC
  • EP2643970B1 patent drawingFigure 1~2
  • EP2643970B1 patent drawingFigure 3~4
  • EP2643970B1 patent drawingFigure 5A~5B

AI summary

An apparatus and method for deriving a motion vector predictor or a motion vector predictor candidate or a motion vector or a motion vector candidate for a current block are disclosed. In video coding systems, the spatial and temporal redundancy is exploited using spatial and temporal prediction to reduce the information to be transmitted or stored. Motion Vector Prediction (MVP) has been used to further conserve the bitrate associated with motion vector coding. The MVP technique being developed for the current HEVC only considers the motion vector having the same reference list and the same reference picture index as the current block to be an available spatial motion vector predictor candidate. It is desirable to develop a MVP scheme that can improve the availability of the motion vector predictor candidate based on motion vectors from the spatially neighboring block. Accordingly, an apparatus and method for determining a motion vector predictor or motion vector predictor candidate or motion vector or motion vector candidate for a current block based on motion vectors associated with reference pictures in list 0 and list 1 of a neighboring block are disclosed. The improved MVP scheme may reduce motion vector residues and consequently the coding efficiency can be improved. Furthermore, the MVP scheme is based on a priority order so that the predictor can be derived at the decoder using decoded information without side information.