MP4 Depth Representation Data Extraction via Sample Entry Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The current methods for storing depth representation information in MP4 files are inefficient, as they redundantly describe decoding unit information and lack direct access to depth representation data, complicating processing for stereoscopic vision generation outside of the decoder.

Innovation Solution

The proposed solution involves sampling and grouping encoded depth image streams with the same depth representation information, allowing for efficient description and direct extraction of depth representation data within the MP4 file, reducing redundancy and simplifying processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If depth representation information is stored in Sample Entry outside the elementary stream in MP4 file, then direct access to depth representation data is enabled, but redundant description of decoding unit information occurs

Engineering Contradiction:
Improvedirect access to depth representation dataVSAvoidredundant description of decoding unit information
Core Design Contradiction:
Ease of operationVSLoss of substance

Solution Approach 1:

The patent extracts only the necessary depth representation information (depth image type, min/max values) from the elementary stream and stores it separately in the MP4 file's Sample Entry. This extraction enables direct access to depth data without requiring extraction of the entire elementary stream, while the patent avoids redundancy by storing only essential parameters rather than complete decoding unit information.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the storage of depth representation information from the elementary stream by placing it in a separate Sample Entry structure in the MP4 file. This segmentation allows independent access to depth representation data while maintaining a clear separation between the elementary stream and metadata, preventing redundant description of decoding unit information.

Inventive Principle:
Principle #1Segmentation

2Loss of information

If depth representation information is extracted from the elementary stream, then complete depth data is obtained, but processing complexity increases

Engineering Contradiction:
Improvecomplete depth dataVSAvoidprocessing complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent extracts only the necessary depth representation information (depth image type, min/max values) from the elementary stream and stores it separately in the MP4 file's Sample Entry. This extraction enables direct access to depth data without requiring extraction of the entire elementary stream, thereby reducing processing complexity while maintaining complete depth data availability.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary extraction and storage of depth representation information during the encoding phase, placing it in the Sample Entry before the elementary stream is fully processed. This preliminary action simplifies subsequent processing by having the depth representation data ready for direct access without requiring complex extraction operations during playback.

Inventive Principle:
Principle #10Preliminary action

3Quantity of substance

If depth representation information is stored in MP4 file, then storage efficiency is improved, but compatibility with existing MP4 structure is reduced

Engineering Contradiction:
Improvestorage efficiencyVSAvoidcompatibility with existing MP4 structure
Core Design Contradiction:
Quantity of substanceVSAdaptability or versatility

Solution Approach 1:

The patent implements the depth representation information storage within the existing MP4 Sample Entry structure, making the Sample Entry serve multiple functions: it continues to store elementary stream reference information while also accommodating depth representation information. This multi-functionality approach maintains compatibility with existing MP4 structures while achieving improved storage efficiency for depth data.

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

Data Source

PatentEP3742747B1Information processing apparatus and information processing method
Publication Date: 2024.02.14 SONY GROUP CORP
  • EP3742747B1 patent drawingFigure 1
  • EP3742747B1 patent drawingFigure 2
  • EP3742747B1 patent drawingFigure 3

AI summary

The present disclosure relates to an information processing apparatus and an information processing method, which enable efficient description of depth representation information in an MP4 file. A file generation unit samples and arranges an elementary stream obtained by encoding a depth image including pixel values, which are obtained by quantizing depth information indicating a position of each pixel in a depth direction, as well as groups samples with the same depth representation information regarding the depth image and generates the MP4 file in which the depth representation information of each group is described. The present disclosure can be applied to, for example, a file generation apparatus and the like.