Multiview Video Encoding Reference Priority Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for combining spatially scalable image encoding and multiview encoding result in high decoding processing loads and low efficiency due to the complexity of inter-layer and inter-view references.

Innovation Solution

A moving picture encoding apparatus that prioritizes inter-layer and inter-view references by using a base layer encoder, enhancement layer encoder, enhancement view base layer encoder, and enhancement view enhancement layer encoder to generate reconstructed images for efficient encoding and decoding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If spatially scalable encoding and multiview encoding are combined, then encoding functionality is improved, but decoding processing load increases and decoding efficiency decreases

Engineering Contradiction:
Improveencoding functionalityVSAvoiddecoding efficiency
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent segments the reference image selection process into distinct layers: base layer and enhancement layer. The segmentation allows the decoder to process references in a hierarchical manner, first establishing base layer references and then building enhancement layer references upon them. This reduces the complexity of simultaneously managing all inter-layer and inter-view references, thereby improving decoding efficiency while maintaining the combined encoding functionality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies preliminary action by pre-determining the priority order of reference images before decoding begins. The encoder establishes a reference priority structure where base layer references are processed first, followed by enhancement layer references. This preliminary organization of reference priorities allows the decoder to systematically process references in the correct order without needing to evaluate all possible references simultaneously, reducing processing load and improving decoding efficiency.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If inter-layer and inter-view references are both used, then encoding quality is improved, but device complexity increases

Engineering Contradiction:
Improveencoding qualityVSAvoidprocessing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the reference management into base layer and enhancement layer components. By dividing the reference structure into these segments, the system maintains high encoding quality through both inter-layer and inter-view references while reducing processing complexity. Each segment handles a specific aspect of reference management, preventing the need to simultaneously process all references in a monolithic structure.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by assigning different priority levels to different reference images based on their layer and view characteristics. Base layer references receive higher priority than enhancement layer references, and this priority differentiation is applied locally to each reference image. This allows the system to maintain high encoding quality by utilizing all references while managing complexity through localized priority assignments rather than uniform processing.

Inventive Principle:
Principle #3Local quality

3Measurement precision

If multiple reference images are processed, then decoding accuracy is improved, but processing time increases

Engineering Contradiction:
Improvedecoding accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-establishing the priority order of multiple reference images before decoding begins. The encoder communicates the reference priority information to the decoder, allowing the decoder to process references in a predetermined optimal order. This preliminary organization enables the decoder to maintain high decoding accuracy by considering multiple references while minimizing processing time through systematic prioritization rather than exhaustive evaluation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent ensures continuity of useful action by organizing reference processing into a continuous hierarchical flow. Base layer references are processed first and continuously available, followed by enhancement layer references that build upon them. This continuous processing approach allows the decoder to maintain high decoding accuracy through multiple references while avoiding interruptions or backtracking, thereby reducing overall processing time compared to non-hierarchical approaches.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentUS10764604B2Moving picture encoding method, moving picture encoding apparatus, moving picture decoding method, and moving picture decoding apparatus
Publication Date: 2020.09.01 SUN PATENT TRUST
  • US10764604B2 patent drawing
  • US10764604B2 patent drawing
  • US10764604B2 patent drawing

AI summary

A spatially scalable-multiview moving picture encoding apparatus includes: a base layer encoder which encodes a base layer; an enhancement layer encoder which performs spatially scalable encoding on an enhancement layer, using a reconstructed image of the base layer generated by the base layer encoder; an enhancement view base layer encoder which performs multiview encoding on the base layer of the enhancement view, using a reconstructed image of the base layer generated by the base layer encoder; and an enhancement view enhancement layer encoder which performs multiview encoding on the enhancement layer of the enhancement view, using a reconstructed image of the enhancement layer generated by the enhancement layer encoder.