Video Bitstream Parsing with Sample Groups for AUD and EOS Placement

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video coding systems face challenges in efficiently integrating scalable bitstreams with different temporal sub-layers or layers into different tracks, leading to complexity in parsing and decoding due to issues with non-video coding layer units like AUD, EOS, and EOB, which require special handling and can complicate the reconstruction process.

Innovation Solution

A file parser and generator are introduced to manage network abstraction layer units, including sample groups that indicate the presence or absence of AUD, EOS, and EOB units, ensuring correct placement and inclusion based on operating points, thereby simplifying the reconstruction process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If scalable bitstreams are split into different tracks for different temporal sub-layers or layers, then users with different capabilities can receive appropriate NAL units, but the parsing and decoding complexity increases due to special handling requirements for non-video coding layer units

Engineering Contradiction:

Improveadaptability to different user capabilitiesVSAvoidparsing and decoding complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The bitstream is segmented into different tracks based on temporal sub-layers or layers, allowing clients to select appropriate tracks according to their capabilities. This segmentation enables adaptive streaming while managing complexity through organized separation of different video quality levels.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Sample groups are used to pre-indicate the presence or absence of non-video coding layer units (AUD, EOS, EOB) in access units. This preliminary information allows parsers to efficiently determine whether to expect and process these special NAL units, reducing the need for complex runtime analysis and special handling.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If non-video coding layer units (AUD, EOS, EOB) are included in scalable bitstreams, then proper delimitation and sequencing are achieved, but the reconstruction process becomes more complex

Engineering Contradiction:

Improvebitstream delimitation and sequencingVSAvoidreconstruction complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The presence of non-video coding layer units is pre-indicated through sample group information, allowing the parser to know in advance whether AUD, EOS, or EOB units will be present in upcoming access units. This eliminates the need for complex runtime detection and handling logic.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Sample groups act as an intermediary mechanism that carries information about the presence of special NAL units. This intermediary structure simplifies the interaction between the encoded bitstream and the parser, providing a clear signaling mechanism without requiring complex parsing logic.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Manufacturing precision

If sample groups are used to indicate presence of non-video coding layer units, then correct placement is ensured, but additional processing steps are required

Engineering Contradiction:

Improveplacement precision of NAL unitsVSAvoidprocessing efficiency

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

Sample groups provide pre-encoded information about the presence and placement of non-video coding layer units. This preliminary information is available during the encoding phase and is directly usable during decoding, eliminating the need for complex runtime analysis to determine unit placement.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The sample group structure is self-contained and carries all necessary information about non-video coding layer unit presence and placement. This self-service mechanism allows the bitstream to be self-descriptive, reducing the need for external control logic or complex processing to determine correct unit placement.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS12519995B2File generator, file parser, methods and video bitstream with improved access unit delimiter, end of sequence, end of bitstream and further non-video coding layer network access layer units

Publication Date: 2026.01.06 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV

AI summary

A file parser for generating an output video bitstream from a file according to an embodiment is provided. The file parser is configured to receive the file being an input video bitstream, wherein the input video bitstream has a video encoded therein. The input video bitstream has a sample group, the sample group having a network abstraction layer unit; and/or the input video bitstream has an indication that an access unit of a plurality of access units of the input video bitstream or a sample of a plurality of samples of the input video bitstream, which does not have the network abstraction layer unit, should have the network abstraction layer unit. The file parser is configured to generate the output video bitstream from the input video bitstream depending on the sample group.