Face Data Acquirer Automates Video Conference Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current face recognition systems require manual user effort to collect and update face data, which is time-consuming and inefficient, and lack an automated method to adapt to changes in appearance over time.
Innovation Solution
A face data acquirer module that automatically captures and processes images from video conferences to extract and store face data, determining storage based on improved detection accuracy, differences in lighting, and image quality, ensuring updated and varied data for enhanced recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual user indication is used to obtain face data, then face recognition accuracy can be maintained, but user time and effort are significantly consumed
Solution Approach 1:
The system automatically performs face data collection, storage, and updating without requiring manual user intervention. The face data acquirer autonomously captures images from video conferences, extracts face data, associates it with identities, and manages the face database, thereby eliminating the need for users to manually indicate face identities while maintaining recognition accuracy.
Solution Approach 2:
The system proactively collects and stores face data in advance during video conference interactions, so that when face recognition is needed, sufficient data is already available. This preliminary automated data collection eliminates the need for users to manually gather face images when recognition is required.
2Quantity of substance
If face data is stored without selection criteria, then data quantity increases, but recognition accuracy may not improve and storage efficiency decreases
Solution Approach 1:
The system incorporates an image analysis module that evaluates each extracted face data against existing data, determining whether to store it based on whether it improves face detection accuracy. This feedback mechanism ensures that only beneficial face data is stored, maintaining high recognition accuracy while avoiding redundant storage.
Solution Approach 2:
The system changes the parameter of face data storage by introducing quality assessment criteria. Instead of storing all extracted face data uniformly, it selectively stores data based on measured differences in lighting conditions, image quality, and contribution to detection accuracy, thereby optimizing both data quantity and recognition performance.
3Adaptability or versatility
If all extracted face data is stored, then data coverage increases, but storage resources are wasted on low-quality or redundant data
Solution Approach 1:
The system applies different quality standards and storage decisions to different face data based on their individual characteristics. High-quality face data with good lighting and clear features are stored, while low-quality or redundant data are discarded. This local quality assessment ensures optimal use of storage resources while maintaining comprehensive data coverage.
4Productivity
If face data is not updated over time, then storage operations are reduced, but face recognition becomes obsolete as appearance changes
Solution Approach 1:
The face data storage system is made dynamic and adaptive, automatically updating face data over time as new images are captured during video conferences. The image analysis module continuously evaluates whether updated face data should replace or supplement existing data, ensuring the face database remains current with appearance changes while maintaining efficient storage operations.
Data Source
AI summary
A face data acquirer includes an image capture module arranged to capture an image from a video stream of a video conference. A face detection module is arranged to determine a subset of the image, the subset representing a face. An identity acquisition module is arranged to acquire an identity of a video conference participant coupled to the face represented by the subset of the image. A face extraction module is arranged to extract face data from the subset of the image and to determine whether to store the extracted face data for subsequent face recognition. A corresponding end user video conference device, server, method, computer program and computer program product are also provided.