Video Parameter Set Parallel Processing Signaling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video coding technologies, such as the HEVC standard and its extensions, face challenges in efficiently managing parallel processing across layers due to complexities in pipelining design, memory cycles, and delay requirements, particularly when using tiles and wavefront parallel processing (WPP), which can introduce delays and require computationally burdensome parsing efforts.

Innovation Solution

The proposed solution involves providing an early indication of parallel processing tools, such as tile configurations and WPP settings, within a video parameter set (VPS) to enable efficient resource allocation and reduce decoding delays. This includes signaling information about tile alignment, loop filters, and entropy coding synchronization, allowing for parallel decoding of video data across layers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If tile configuration and wavefront parallel processing are used for parallel processing across layers, then processing speed and resource utilization are improved, but decoding delays increase and parsing complexity becomes computationally burdensome

Engineering Contradiction:
Improveparallel processing efficiencyVSAvoiddecoding delay
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by signaling tile configuration and parallel processing information in the VPS (Video Parameter Set) before the actual decoding process. This allows the decoder to prepare parallel processing structures and allocate resources in advance, reducing actual decoding delays while maintaining parallel processing efficiency.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If tile configuration and wavefront parallel processing are used for parallel processing across layers, then processing speed and resource utilization are improved, but device complexity and parsing efforts increase

Engineering Contradiction:
Improveparallel processing efficiencyVSAvoidparsing complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

By signaling parallel processing information in the VPS beforehand, the complex parsing and configuration setup is performed once during initialization rather than repeatedly during decoding. This reduces real-time parsing complexity while maintaining parallel processing capabilities.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The VPS structure serves multiple functions: it not only contains standard video parameters but also signals tile configuration and parallel processing information. This multi-functionality reduces the need for separate signaling mechanisms, thereby reducing overall device complexity.

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

3Device complexity

If parallel processing information is signaled late in the decoding process, then device complexity is reduced, but decoding delays increase and resource allocation efficiency decreases

Engineering Contradiction:
Improvesignaling structure simplicityVSAvoiddecoding delay
Core Design Contradiction:
Device complexityVSLoss of time

Solution Approach 1:

The patent resolves this contradiction by moving the signaling of parallel processing information to the VPS, which is decoded early in the process. This preliminary signaling enables timely resource allocation and parallel processing setup without adding significant device complexity, as the VPS is a standard structure already present in HEVC.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3058729B1Indication of parallel processing in video coding
Publication Date: 2020.08.26 QUALCOMM INC
  • EP3058729B1 patent drawingFigure 1
  • EP3058729B1 patent drawingFigure 2
  • EP3058729B1 patent drawingFigure 3

AI summary

In an example, a method of decoding video data includes decoding, from a video parameter set (VPS) of a multi-layer bitstream, data that indicates at least one of a tile configuration for layers of the multi-layer bitstream or a parallel processing configuration for layers of the multi-layer bitstream. The method also includes decoding the multi-layer bitstream in accordance with the data decoded from the VPS.