Widespread use of face recognition systems: even face recognition when wearing a mask

Face recognition is becoming more common these days, but in the past it was used only for limited purposes, such as identifying a person's face.

However, today it is used in various situations such as customer data analysis, digital signage, attendance management, ATM / mobile banking biometric authentication, parking lot and locker keys, etc. Furthermore, there is an increasing need to implement these functions in embedded equipment/edge devices. Recently, we have received more inquiries about mask detection and face recognition when wearing a mask as a countermeasure against after-corona and with-corona.

Therefore, in this article, after touching on the hardware and face recognition software that are most suitable for building face recognition systems in embedded devices, we will introduce examples of mask detection and face recognition when wearing a mask, along with videos, so please take a look. .

Optimal hardware for facial recognition in embedded devices

Depending on the application, embedded devices are often required to be able to achieve high-speed processing with low power consumption. IEI Integration's products are the most suitable products to meet such demands.

IEI Integration provides Intel® Movidius™ and FPGA deep learning accelerators and gateway terminals to meet customer's requirements for embedded devices.

​Click here for details on IEI Integration.

Highly accurate AI face recognition engine FaceMe®

CyberLink's FaceMe® is AI facial recognition software for embedded devices such as edge devices. It is possible to add a highly accurate face recognition function to mobile devices, embedded devices, IoT, servers, etc.

From here, we will introduce the features and benefits of FaceMe®.

AI face recognition engine with high recognition accuracy and safety

CyberLink's face recognition engine FaceMe®, which employs advanced AI technology based on deep learning, ranks among the top in the world in face recognition technology benchmark tests (VISA, WILD tests) conducted by the National Institute of Standards and Technology (NIST). increase.

FaceMe® has world-class facial recognition speed and accuracy, and can recognize faces even if the face is tilted up to 60 degrees.

In addition, as a countermeasure against spoofing using photos and images, we will realize a safer biometric authentication environment by combining multiple biometric authentication technologies that prevent spoofing.
FaceMe® supports not only 2D cameras built into smartphones and tablets, but also 3D cameras like Intel® RealSense™ to prevent spoofing.

Flexible support from edge IoT to cloud servers

Easily build complex systems where multiple OSs and devices coexist, such as door security systems, digital signage, kiosk terminals, POS systems, and other devices and IP cameras for detection and aggregation on a cloud server. is possible.

 

FaceMe® product line

FaceMe® offers three product groups:

Mask detection/face recognition when wearing a mask

As a countermeasure against after-corona and with-corona, it is necessary to wear masks in various situations, and the demand for whether or not to wear a mask and facial recognition when wearing a mask is growing. , Whether or not a mask is worn, and facial recognition when wearing a mask can be performed.

The following two-minute video introduces whether or not a person is wearing a mask, and how facial recognition is performed when wearing a mask. Please take a look.

Document download

In this article, we introduced the best hardware and face recognition software for building a face recognition system on the edge.

I want to know more about this product! For those who say, there is a document that explains a little more details, so please download it from the following.

Inquiry

I want to ask a question directly to the person in charge! I want an explanation! If so, please feel free to contact us using the button below.