Integrating Scene Index Data with Video Streams
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing picture processing technologies lose attribute information about scenes in moving picture data when transferring data between different apparatuses, making it difficult to efficiently navigate and reproduce specific scenes.
Innovation Solution
A picture processing apparatus and method that detects attribute information, creates index data, and integrates it with moving picture data as stream data, ensuring that index data is stored and retained even after data transfer, allowing for efficient scene navigation and reproduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If independent management information files are used to store scene information, then scene indexing capability is improved, but data reliability deteriorates due to file loss during transfer
Solution Approach 1:
The patent merges the management information file containing scene attribute information with the moving picture data stream, creating an integrated stream data structure. This combination ensures that when moving picture data is transferred, the scene indexing information travels with it, eliminating the file loss problem while maintaining scene navigation capability.
Solution Approach 2:
The patent performs preliminary detection of attribute information (such as face detection results) during the image acquisition phase and stores this information in advance within the stream data structure. This preliminary action ensures that scene indexing data is already prepared and embedded before transfer occurs, eliminating the need for separate management files.
2Ease of operation
If thumbnails are created from different parts of moving picture data, then scene navigation is improved, but data volume increases
Solution Approach 1:
The patent extracts only the essential attribute information (such as face detection results, scene identifiers, and position data) from the moving picture data, rather than storing complete thumbnails. This extraction approach provides sufficient scene navigation capability while significantly reducing the additional data volume compared to full thumbnail storage.
Data Source
AI summary
Disclosed herein is a picture processing apparatus including: an attribute information detection section configured to detect attribute information about a plurality of pictures included in moving picture data; an index data creation section configured to create index data for indexing the attribute information to the positions in the moving picture data of those of the plurality of pictures which correspond to the attribute information; and a stream data creation section configured to create stream data including the moving picture data and the index data.


