Face Recognition Indexing for Image Search Bottlenecks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital image processing devices do not provide an efficient way for users to determine if an image of a specific person is stored, the number of times the person has been photographed, and which images include the person alone or with others.
Innovation Solution
A digital image processing apparatus equipped with a face recognition unit, face index assignment unit, and relation index update unit that recognizes faces, assigns unique indices, and updates relation indices based on co-occurrence in images, allowing users to easily identify and sort images by face presence and intimacy grades.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional digital image processing apparatus stores images without face recognition, then storage capacity is maintained, but user ability to determine face information is poor
Solution Approach 1:
The face recognition unit performs face recognition and extracts face information during the image storage process, before the user needs to query for face information. This preliminary action ensures that face information is already processed and stored in an accessible format, eliminating the need for users to manually examine each image to determine face content.
2Ease of operation
If face recognition is implemented to capture face information, then user convenience is improved, but processing time increases
Solution Approach 1:
Face recognition and information extraction are performed automatically when images are stored, rather than when users query for face information. This shifts the processing time to the storage phase, making subsequent user operations instantaneous and convenient without requiring users to wait for processing during their queries.
Solution Approach 2:
The system automatically performs face recognition and information extraction without requiring user intervention or initiation. The apparatus serves itself by automatically processing images during storage, eliminating the need for users to manually analyze images or initiate processing operations.
3Loss of information
If detailed face information is stored for each image, then information completeness is improved, but storage requirements increase
Solution Approach 1:
Instead of storing complete image data multiple times for face information, the system extracts and stores only the essential face information (such as face indices, relationship indices, and presence/absence data) as separate metadata. This copying approach provides complete face information while requiring minimal additional storage space compared to storing full image copies.
Data Source
AI summary
A digital image processing apparatus for easily obtaining information regarding a face, a method of controlling the same, and a recording medium for storing a program for executing the method. The digital image processing apparatus includes a face recognition unit recognizing a face from an image, a face index assignment unit assigning a face index to the recognized face if the face recognized from the image is not a previously recognized face, and a relation index update unit that, if the image includes a second face, updates a relation index corresponding to the number of times the first face and second face appear together in a plurality of images.