Inter-layer Prediction Type Signaling in MV-HEVC

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In multi-layer video coding, the existing techniques face inefficiencies in signaling inter-layer dependency types, leading to increased overhead due to the need to code multiple instances of dependency type syntax elements, especially in bitstreams with a large number of layers.

Innovation Solution

The proposed solution involves signaling a single syntax element to indicate whether the dependency type for each layer combination is signaled, allowing the decoder to assume a predetermined type if not explicitly signaled, thereby reducing the need for multiple instances of dependency type syntax elements and minimizing signaling overhead.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If dependency type syntax elements are coded for each layer combination, then the inter-layer prediction accuracy is improved, but the signaling overhead increases

Engineering Contradiction:
Improveinter-layer prediction accuracyVSAvoidsignaling overhead
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent merges the dependency type information for multiple layer combinations into a single syntax element. Instead of coding separate dependency type syntax elements for each layer combination, the encoder signals one syntax element that represents the dependency type for all layer combinations, thereby reducing signaling overhead while maintaining prediction accuracy

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The single syntax element serves multiple functions by indicating the dependency type for all layer combinations simultaneously. This universal syntax element allows the decoder to determine the inter-layer prediction type for any layer combination without requiring separate signaling, thus reducing overall signaling overhead

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Manufacturing precision

If multiple instances of dependency type syntax elements are signaled, then the coding precision for each layer is improved, but the bitstream complexity increases

Engineering Contradiction:
Improvecoding precisionVSAvoidbitstream complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent combines multiple dependency type syntax elements into a single syntax element in the bitstream. This merging reduces bitstream complexity by eliminating redundant signaling while preserving the necessary coding precision through the unified syntax element that applies to all layer combinations

Inventive Principle:
Principle #5Merging (Combining)

3Measurement precision

If dependency type syntax elements are signaled for all layer combinations, then the decoding accuracy is improved, but the processing time increases

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

Solution Approach 1:

The patent merges dependency type information for all layer combinations into a single syntax element, allowing the decoder to determine inter-layer prediction types for all layers simultaneously. This reduces processing time by avoiding multiple separate syntax element interpretations while maintaining decoding accuracy through the unified dependency type information

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP3050299B1Inter-view dependency type in MV-hevc
Publication Date: 2019.12.25 QUALCOMM INC
  • EP3050299B1 patent drawingFigure 1
  • EP3050299B1 patent drawingFigure 2
  • EP3050299B1 patent drawingFigure 3

AI summary

A video decoder receives a value for a first syntax element representing whether a dependency type syntax element for a current layer is signaled, wherein the dependency type syntax element identifies a type of dependency of a current layer relative to a reference layer; and in response to the value for the first syntax element indicating that the dependency type syntax element is not signaled determines that the type of dependency of the current layer relative to the reference layer is a predetermined type and decodes a block of the current layer using inter-layer prediction conforming to the predetermined type.