Video Parameter Set Segmentation for HEVC Signaling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video coding standards, such as HEVC, face challenges in efficiently processing and transmitting video parameter sets, particularly in decoupling infrequently changing information from coded video block data, which can lead to increased bandwidth and complexity in network elements like MANEs.

Innovation Solution

The introduction of an offset syntax element in video parameter sets allows network elements to skip variable length coded syntax elements, reducing processing complexity and enabling hierarchical formatting, while also including session negotiation and HRD parameters directly in the VPS to reduce signaling overhead and facilitate future extensions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of substance

If variable length coded syntax elements are included in video parameter sets, then compression efficiency is improved, but processing complexity in network elements increases

Engineering Contradiction:
Improvebandwidth consumptionVSAvoidprocessing complexity in network elements
Core Design Contradiction:
Loss of substanceVSDevice complexity

Solution Approach 1:

The video parameter set is segmented into fixed length coded syntax elements and variable length coded syntax elements. Network elements can process only the fixed length portion without entropy decoding, while entropy decoding is performed only for the variable length portion, thereby reducing processing complexity while maintaining compression efficiency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Fixed length coded syntax elements are extracted and placed before variable length coded syntax elements in the video parameter set. This allows network elements to extract and process the fixed length portion independently without requiring full entropy decoding, reducing their processing burden.

Inventive Principle:
Principle #2Taking out (Extraction)

2Device complexity

If infrequently changing information is coupled with coded video block data, then signaling is simplified, but bandwidth consumption increases

Engineering Contradiction:
Improvesignaling complexityVSAvoidbandwidth consumption
Core Design Contradiction:
Device complexityVSLoss of substance

Solution Approach 1:

Infrequently changing information (fixed length coded syntax elements) is extracted and separated from the variable length coded syntax elements. This allows network elements to process the fixed length portion without entropy decoding, reducing bandwidth consumption while keeping signaling simple.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The parameter set structure is organized into different dimensions: fixed length coded elements are placed in a separate dimension (before variable length elements) that can be processed independently, while variable length elements remain in the original dimension requiring entropy decoding. This dimensional separation reduces bandwidth consumption.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Reliability

If entropy decoding is performed in network elements, then complete parameter set processing is achieved, but processing time increases

Engineering Contradiction:
Improveprocessing completenessVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The parameter set is segmented such that network elements can process the fixed length coded portion completely without entropy decoding, achieving partial processing completeness quickly. Entropy decoding is then performed only on the variable length portion, reducing total processing time while maintaining necessary completeness.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Fixed length coded syntax elements are placed before variable length coded elements, allowing network elements to perform preliminary processing of the fixed length portion without entropy decoding. This preliminary action reduces the amount of data requiring time-consuming entropy decoding later.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3200465B1Video parameter set for HEVC and extensions
Publication Date: 2023.12.13 QUALCOMM INC
  • EP3200465B1 patent drawingFigure 1
  • EP3200465B1 patent drawingFigure 2
  • EP3200465B1 patent drawingFigure 3

AI summary

A video coder can be configured to receive in a video parameter set, one or more syntax elements that include information related to hypothetical reference decoder parameters; receive in the video data a first sequence parameter set comprising a first syntax element identifying the video parameter set; receive in the video data a second sequence parameter set comprising a second syntax element identifying the video parameter set; and, code, based on the one or more syntax elements, a first set of video blocks associated with the first parameter set and second set of video blocks associated with the second parameter set.