HEVC High-Level Syntax Extensions for Multiview Video Coding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video coding standards, such as HEVC, face challenges in efficiently supporting extensions like multiview and 3D video coding with high-level syntax changes, particularly in managing motion vector predictions between different types of reference pictures.

Innovation Solution

The implementation of specific techniques within the HEVC base specification and its extensions, including disabling motion vector prediction between short-term and long-term reference pictures, and using hooks to manage prediction types, ensures compatibility and efficient coding for multiview and 3D video applications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If motion vector prediction is enabled between all reference pictures to improve compression efficiency, then compression efficiency improves, but coding complexity and error propagation increase when mixing short-term and long-term reference pictures

Engineering Contradiction:
Improvecompression efficiencyVSAvoidcoding complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments reference pictures into two distinct categories: short-term reference pictures and long-term reference pictures. This segmentation allows the system to apply different motion vector prediction rules to each category, simplifying the coding process by preventing complex cross-category predictions while maintaining compression efficiency through category-appropriate prediction strategies

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of enabling motion vector prediction between all reference picture types (the conventional approach), the patent inverts the logic by explicitly disabling predictions between short-term and long-term reference pictures. This inversion simplifies the prediction model by eliminating problematic cross-category predictions while still allowing efficient predictions within each category

Inventive Principle:
Principle #13The other way round (Inversion)

2Adaptability or versatility

If high-level syntax extensions are added to support multiview and 3D video coding, then functionality and adaptability improve, but compatibility with base HEVC specification and implementation complexity worsen

Engineering Contradiction:
Improvefunctionality for multiview and 3D codingVSAvoidimplementation complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent creates a universal motion vector prediction framework that serves multiple functions: it handles conventional single-view video coding, multiview video coding, and 3D video coding through the same basic mechanisms. By making the prediction model view-independent and extension-friendly, the system achieves high adaptability without proportionally increasing implementation complexity

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

Solution Approach 2:

The patent enables extensions to HEVC by allowing changes at the parameter level (such as adding new reference picture types and modifying prediction rules) rather than requiring fundamental changes to the coding framework. This approach maintains compatibility with the base specification while enabling advanced functionalities through controlled parameter modifications

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2826248B1High-level syntax extensions for high efficiency video coding
Publication Date: 2021.11.24 QUALCOMM INC
  • EP2826248B1 patent drawingFigure 1
  • EP2826248B1 patent drawingFigure 2
  • EP2826248B1 patent drawingFigure 3

AI summary

In one example, a device includes a video coder configured to code a picture order count (POC) value for a first picture of video data, code a second-dimension picture identifier for the first picture, and code, in accordance with a base video coding specification or an extension to the base video coding specification, a second picture based at least in part on the POC value and the second-dimension picture identifier of the first picture. The video coder may comprise a video encoder or a video decoder. The second-dimension picture identifier may comprise, for example, a view identifier, a view order index, a layer identifier, or other such identifier. The video coder may code the POC value and the second-dimension picture identifier duing coding of a motion vector for a block of the second picture, e.g., during advanced motion vector prediction or merge mode coding.