Picture Header Signaling for Accurate Video Picture Boundaries

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video coding technologies face challenges in accurately determining the starting point of a picture within a multi-layer bitstream, leading to potential coding errors and suboptimal codec performance.

Innovation Solution

Ensuring that a picture unit (PU) includes only one picture header (PH) and one picture, with the layer identifier (ID) of the PH network abstraction layer (NAL) unit equal to the layer ID of the associated video coding layer (VCL) NAL unit and the temporal ID of the PH NAL unit equal to the temporal ID of the PU, allowing precise determination of the picture starting point without coding errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multi-layer bitstream structure is used to support scalable video coding, then video quality and adaptability are improved, but determining the starting point of pictures becomes complex and error-prone

Engineering Contradiction:

Improvescalable video codingVSAvoidpicture starting point identification

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent segments the bitstream into distinct picture units (PU), each containing exactly one picture header NAL unit and one associated VCL NAL unit. This segmentation ensures that each PU is an independent, self-contained entity that can be reliably identified and processed without ambiguity about picture starting points, thereby resolving the reliability issue while maintaining multi-layer scalability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces the picture unit (PU) as an intermediary structure between the multi-layer bitstream and the decoding process. The PU acts as a mediator that organizes and delimits picture data, making the starting points of pictures explicitly identifiable through its structure (PH_NAL unit followed by associated VCL NAL unit), thus solving the picture identification problem in scalable video coding.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If multiple picture headers are allowed in a picture unit, then flexibility in coding structure is improved, but coding errors increase and performance deteriorates

Engineering Contradiction:

Improvecoding structure flexibilityVSAvoidcoding accuracy

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The patent applies local quality by imposing a specific constraint at the picture unit level: exactly one picture header NAL unit per PU. This local restriction ensures coding accuracy and eliminates ambiguity about picture starting points, while the overall system maintains flexibility through the multi-layer structure and the ability to have multiple PUs. Each PU becomes a locally optimized unit with precise structure.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS12495162B2Signaling of picture header in video coding

Publication Date: 2025.12.09 HUAWEI TECH CO LTD

AI summary

A method of decoding is provided. The method includes receiving a video bitstream including a picture unit (PU), the PU including a picture header (PH) network abstraction layer (NAL) unit having a picture header NAL unit type (PH_NUT) and an associated video coding layer (VCL) NAL unit, the PH_NUT signifying that a layer identifier (ID) of the PH NAL unit is equal to a layer ID of the associated video coding layer (VCL) NAL unit and a temporal ID of the PH NAL unit is equal to a temporal ID of the PU; and decoding a coded picture from the PU to obtain a decoded picture.