Inter-Prediction Candidate List Construction for Video Coding Efficiency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The existing moving picture coding schemes, such as MPEG-4 AVC/H.264, face challenges in reducing the coding amount of coding information for motion-compensated prediction, particularly in scenarios where processing capability is limited.

Innovation Solution

A moving picture coding device that derives and codes inter-prediction information candidates from neighboring or temporally different prediction blocks, constructs an inter-prediction information candidate list, and supplements candidates with predetermined values to ensure a designated number of candidates are available for motion-compensated prediction, thereby improving coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If motion-compensated prediction is performed using multiple reference pictures with reference indices, then prediction accuracy is improved, but coding amount increases

Engineering Contradiction:

Improveprediction accuracyVSAvoidcoding amount

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent pre-constructs a candidate list of inter-prediction information including motion vectors and reference picture indices before the actual prediction process. This preliminary organization allows the encoder to efficiently select from pre-prepared candidates rather than generating and evaluating all possible combinations during encoding, thus reducing the coding amount while maintaining prediction accuracy

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the parameter representation by encoding motion vector differences (MVD) rather than absolute motion vectors, and by using compact reference picture index representations. This parameter transformation reduces the number of bits required to represent the same prediction information, thereby reducing coding amount while preserving prediction accuracy

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If bi-prediction using L0 and L1 prediction modes is employed, then prediction precision is improved, but device complexity increases

Engineering Contradiction:

Improveprediction precisionVSAvoiddevice complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the bi-prediction process into independent L0 prediction and L1 prediction components, each with its own candidate list and selection mechanism. This segmentation allows the decoder to implement bi-prediction by independently processing two simpler prediction streams rather than managing a single complex bidirectional prediction process, thereby reducing device complexity while maintaining prediction precision

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent uses identical candidate list construction and selection algorithms for both L0 and L1 prediction modes, copying the same computational framework and data structures. This copying approach allows the system to leverage the same optimized code paths and processing logic for both prediction directions, reducing the overall device complexity compared to implementing two different prediction mechanisms

Inventive Principle:

Principle #26Copying

3Productivity

If direct mode is used to construct inter-prediction information from coded/decoded blocks, then coding efficiency is improved, but adaptability decreases

Engineering Contradiction:

Improvecoding efficiencyVSAvoidadaptability

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent implements a dynamic candidate list construction mechanism that adapts the inter-prediction information candidates based on the actual content of neighboring blocks and reference pictures. Rather than using a fixed direct mode template, the system dynamically selects and adjusts candidates from the constructed list, maintaining coding efficiency while improving adaptability to different scene conditions

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent incorporates feedback from the encoding process itself by using the constructed candidate lists to guide the selection of inter-prediction information. The system evaluates multiple candidates from the list and selects the one that provides the best prediction, using this feedback to improve both coding efficiency and adaptability compared to rigid direct mode approaches

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS12167023B2Moving picture coding device, moving picture coding method, moving picture coding program, moving picture decoding device, moving picture decoding method, and moving picture decoding program

Publication Date: 2024.12.10 JVC KENWOOD CORP

AI summary

A prediction information deriving unit derives the inter-prediction information candidates from inter-prediction information of a prediction block neighboring to a coding target prediction block or a prediction block present at the same position as or near the coding target prediction block in a coded picture at a temporally different position from the coding target prediction block. A candidate supplementing unit supplements inter-prediction information candidates having the same prediction mode, reference index, and motion vector until the number of inter-prediction information candidates reaches the designated number of candidates when the number of inter-prediction information candidates is smaller than the designated number of candidates.