Face recognition is an automated process for comparing faces. It relies on cutting-edge computer vision algorithms that detect human faces in images, videos, or live feeds and match them against a database of faces.

Face recognition helps automate tasks and processes, and provide better, more personalized user experiences.

It’s used across various industries, such as security, automotive , healthcare, retail, and more. The most common use cases include device and access control, attendance tracking, ID matching, driver recognition, etc.

Besides security, face recognition also helps facilitate personalization. Being able to recognize individuals in real time helps tailor products, services, and environments to their needs and preferences.

You can easily integrate FaceRecognition into your existing solution or rely on our in-house research and development team to create a custom solution or specialized features for you.

Fast. We typically match faces in a fraction of a second. This speed can be affected by image size, the number of faces in the image, and your specific use case.

Recently, visage|SDK was included in the independent NIST FR Vendor test (FRVT). Our FaceRecognition was the fastest and most accurate lightweight algorithm compared with all other submissions in the report. It stood out with an execution time of only 26.25 ms and memory usage of just 73 MB. Among all the lightweight submissions, it achieved the highest accuracy on unconstrained images with a False Positive Rate of around 0,0005%. These are images taken ‘in the wild’, without controlled conditions.

Read the full report here.

FaceRecognition is available for a wide range of platforms including desktop, mobile, web, and embedded devices. It can process a variety of file formats, including photos, videos, and live streams. Because of its extremely lightweight nature, it’s the perfect fit for edge devices.

FaceRecognition runs entirely on the client-side. There are no cloud API calls, no data transfers, and no need for a network connection to the device. Instead of sending data – such as facial images – to a server through an internet connection, FaceRecognition compares facial features from an image to the face descriptors already saved in an offline database.

That means that all the processed data stays on-device, without any intermediaries or third parties having access to it.

Liveness detection helps prevent spoofing attempts designed to deceive a face recognition system using a substitute for another person’s face (usually their photo, video recording or a 3D mask). If the spoofing attack succeeds, the fraudster acquires the privileges or access rights of another person.

Liveness detection prompts the user to perform a simple and quick facial action, such as a smile or a blink, then verifies whether the action was performed. This ensures that there’s truly a live person in front of the camera.

Sure! You can try out a simple HTML5 demo here. To explore FaceRecognition in more detail on your preferred platform, get in touch and request your free evaluation license.