Storage Apparatus Metadata Segmentation for Face Data Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing metadata storage technique for face detection, as described in Patent Document 1, is inefficient as it stores metadata in a single structure, requiring time-consuming analysis to extract necessary data and is hindered by varying data sizes due to the variable number of detected human faces, making quick access difficult.
Innovation Solution
A storage apparatus and method that organizes metadata with first metadata for each content, second metadata associated with each face image, and storage location information in a single data set, allowing for efficient storage and retrieval by using continuous storage areas for sets of metadata and face image data, including version information, face location, and number of face images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If metadata is stored as a single structure in a storage area, then storage organization is simple, but reading necessary data takes a long time
Solution Approach 1:
The patent divides the metadata storage structure into multiple separate storage areas: a first storage area for content-level metadata and a second storage area for face-level metadata. This segmentation allows the system to read only the necessary metadata from the appropriate storage area, significantly reducing data reading time while maintaining organized storage structure.
2Adaptability or versatility
If metadata is stored in XML format with nested structures, then data organization is comprehensive, but analysis time to extract target data increases
Solution Approach 1:
The patent segments metadata into distinct storage areas with specific purposes: content-level metadata in the first storage area and face-level metadata in the second storage area. This eliminates the need to parse through nested XML structures to find specific data, as the segmentation inherently organizes data by access patterns and importance.
Solution Approach 2:
The patent extracts face-level metadata from the overall metadata structure and places it in a separate second storage area. This extraction allows the system to quickly access face-related information without analyzing the entire metadata structure, significantly reducing data extraction time while preserving comprehensive metadata organization.
3Adaptability or versatility
If metadata size varies with number of detected faces, then data storage is flexible, but quick access becomes difficult
Solution Approach 1:
The patent segments metadata storage into a first storage area for fixed-size content-level metadata and a second storage area for variable-size face-level metadata. This segmentation maintains storage flexibility to accommodate varying numbers of detected faces while enabling quick access by allowing the system to read only the required portion from the appropriate storage area.
Data Source
AI summary
A storage apparatus and method, a program, and a playback apparatus and method, capable of quickly reading a specific part of data among metadata including metadata associated with faces. A storage controller controls storing face metadata in a storage medium, wherein the face metadata includes a content data set added for each content, content data storage location information indicating the storage location of the content data set, a detected face data set associated with each of face images detected from a content, and detected face data storage location information indicating the storage location of the detected face data set, and wherein the face metadata is configured such that the content data storage location information and face block storage location information indicating the storage location of the detected face data storage location information are described in a single data set. The present invention is applicable to a digital camera.