Server-Client Faceprint Delivery for Mobile Facial Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing facial recognition technologies face challenges in efficiently processing facial images on mobile devices due to computational complexity, especially when dealing with variations in brightness, positioning, and expression, as well as large datasets, which results in high processing power requirements and long recognition times.
Innovation Solution
A system where a server optimizes facial image recognition by delivering a subset of faceprints to the client device, reducing the number of comparisons needed and using computationally less intensive calculations, and selects reference images based on uniqueness, allowing for faster and more efficient recognition on mobile devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If comprehensive facial recognition comparisons are performed on mobile devices, then recognition accuracy is improved, but processing time and computational complexity increase significantly
Solution Approach 1:
The system performs preliminary actions by pre-computing and storing faceprints (compressed facial representations) during a setup phase. These pre-processed faceprints are then used for rapid comparison during actual recognition operations, eliminating the need for complex real-time processing on mobile devices while maintaining high accuracy.
Solution Approach 2:
The system extracts essential facial features and compresses them into compact faceprint representations. By taking out only the most discriminative features and storing them in a compressed format, the system enables fast comparisons without requiring full-resolution image processing, thus reducing processing time while preserving recognition accuracy.
2Measurement precision
If comprehensive facial recognition comparisons are performed on mobile devices, then recognition accuracy is improved, but device complexity and power consumption increase
Solution Approach 1:
The system extracts and stores only the essential facial features in compressed faceprint format, removing unnecessary computational complexity. This extraction approach allows mobile devices to perform simple comparisons rather than complex full-image analyses, significantly reducing device requirements while maintaining accuracy.
Solution Approach 2:
The system creates simplified copies of facial data in the form of faceprints. These compressed representations serve as substitutes for full-resolution images, enabling accurate recognition with minimal computational resources on mobile devices.
3Measurement precision
If comprehensive facial recognition comparisons are performed on mobile devices, then recognition accuracy is improved, but energy consumption increases
Solution Approach 1:
The system performs energy-intensive processing in advance by pre-computing faceprints during setup. This preliminary action shifts the computational burden away from the mobile device during actual use, enabling accurate recognition with minimal power consumption on battery-powered devices.
Solution Approach 2:
The system extracts only the essential facial information needed for recognition and stores it in compressed form. This extraction eliminates the need for energy-intensive processing of full-resolution images during recognition operations, significantly reducing power consumption while maintaining accuracy.
Data Source
AI summary
A server determines a plurality of faceprints representing a plurality of users to be recognized at a client device. Each faceprint contains a number of reference images for a given user that are used to recognize facial images of the user detecting in media captured at the client device. The faceprints delivered to the client device are determined for the client device based on the users likely to be detected in images captured at the client device. The reference images with a given faceprint delivered to the client device are selected by the server based on their recognition value in identifying the users likely to be detected in images captured at the client device.