Video Coding Offset Delay Parameters for Parallel Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video coding techniques face inefficiencies in parallel processing of multi-layer video coding, particularly in signaling offset delay information and determining collocated blocks, which restricts the sharing of information across layers and impacts coding efficiency.

Innovation Solution

Incorporating offset delay information into a video parameter set (VPS) to allow separate specification for each reference layer, and determining the location of collocated blocks by adjusting reference sample locations within layer boundaries, enabling efficient inter-layer prediction and parallel processing across layers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If offset delay information is signaled separately for each layer, then inter-layer prediction accuracy is improved, but signaling complexity increases

Engineering Contradiction:
Improveinter-layer prediction accuracyVSAvoidsignaling complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines offset delay information signaling with the VPS structure, merging layer-specific parameters into a unified parameter set that handles multiple layers. This allows efficient signaling of offset delays for different reference layers without requiring separate signaling mechanisms for each layer, thus improving inter-layer prediction accuracy while controlling signaling complexity.

Inventive Principle:
Principle #5Merging (Combining)

2Productivity

If parallel decoding of multiple layers is implemented, then processing speed is improved, but determination of collocated blocks becomes more complex

Engineering Contradiction:
Improveprocessing speedVSAvoidcollocated block determination complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary determination of collocated block locations by adjusting reference sample locations within layer boundaries before parallel decoding begins. By pre-calculating and storing the adjusted reference sample locations for each layer, the system enables parallel decoding to proceed without complex real-time calculations, thus improving processing speed while managing the complexity of collocated block determination.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If reference sample locations are adjusted within layer boundaries, then parallel processing efficiency is improved, but information sharing across layers is restricted

Engineering Contradiction:
Improveparallel processing efficiencyVSAvoidinformation sharing across layers
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent applies local quality adjustment by modifying reference sample locations specifically within layer boundaries where needed for parallel processing, while preserving the original reference relationships across layers. This localized adjustment ensures that parallel processing efficiency is improved in specific regions without compromising the overall information sharing capability across different layers, as the adjustment is applied selectively rather than globally.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3022925B1Cross-layer parallel processing and offset delay parameters for video coding
Publication Date: 2022.09.21 QUALCOMM INC
  • EP3022925B1 patent drawingFigure 1
  • EP3022925B1 patent drawingFigure 2
  • EP3022925B1 patent drawingFigure 3

AI summary

In an example, a method of coding video data includes determining a location of a reference sample associated with a reference picture of video data based on one or more scaled offset values, where the reference picture is included in a first layer of a multi-layer bitstream and the one or more scaled offset values indicate a difference in scale between the first layer and a second, different layer. The method also includes determining a location of a collocated reference block of video data in the first layer based on the location of the reference sample, and coding a current block of video data in the second layer relative to the collocated reference block.