The Smart ID Engine face verification module provides a robust and accurate solution to match the photo on a user’s ID with a selfie on a mobile device or using remote verification channels. Automatic face matching allows you to improve the quality and security of issuing loans, insurance or bank cards, opening bank accounts, registering for government services in mobile applications and on the web. In addition to document liveliness and document forgery detection, the face verification module helps to prevent forgery in a seamless personal identification process.
The Smart ID Engine face verification module does not detect biometric features, does not create robust face descriptors, does not save or transfer biometric images. Thanks to our data synthesis algorithms, face verification does not contain discrimination with regards to gender, race, or national origin, and is also resistant to changes in illumination and distortions due to uncontrolled capturing conditions.
Data synthesis helps avoid discrimination problems in training (bias), which occurs in many cases where natural unbalanced samples are used. Due to the fact that the system does not detect any biometric features, face verification is close to the way a human does it. The absence of robust face descriptors does not allow clustering or recognizing faces from the database. Thus, the authentication process has become secure for both users and companies.
Our face verification SDK is part of the secure Smart ID Engine system and operates independently, face verification process is performed automatically on end-user devices without any data transfer. The software does not require high-performance computing power, so it can be run even on devices with limited computing power and energy. The solution supports smartphones, tablets, thin clients, portable data terminals, mobile cash registers, smart cameras, etc. The Smart ID Engine face verification module can operate under various operating systems (OS Windows, Linux, Android, iOS) and hardware platforms (x86_64, ARM, MIPS).
The software does NOT transfer personal and biometric data to any third-party services and/or third parties for manual processing, does NOT save or store it: the processing is carried out in the device’s local RAM, and it does NOT require internet access. All the solutions are HIPAA, GDPR, CCPA-compliant.
Green AI-based computer vision technologies developed by our scientists and engineers allow us to minimize the carbon footprint of AI technologies. Our training technology for compact low-bit neural network models performs the face verification process in real time, optimally utilizing the available computing resources. Careful algorithm optimization and implementation make it possible to use devices with low computational performance for secure and efficient verification.
Smart ID Engine SDK allows you to turn an iOS or Android-based mobile device into a verification system thanks to a unified document recognition system and face verification. Our original AI algorithms perform high-speed and autonomous face verification processes. Our algorithm co-optimization for face detection and verification in selfies and documents allows us to quickly, precisely and securely carry out the verification process on a wide range of mobile devices.
Due to the WebAssembly (WASM) support, the Smart ID Engine face verification solution can be implemented in all up-to-date devices with installed browsers (Mozilla Firefox, Google Chrome, Chromium, Microsoft Edge and Apple Safari), including smart TVs and consoles. The software does not require special applications or special plugins. The use of original neural network architectures allows you to process video data obtained from low-end webcams with a resolution of 640×480 pixels. It is important to stress that images and data are not transferred to third parties or services.
Deep software and hardware optimization of the algorithmic recognition principles for various processor architectures allows Smart ID Engine to autonomously perform face verification on any devices with a camera and a processor. Processor architectures MIPS, ARM, x86_64, and a wide range of operating systems (Linux OS, Windows OS) are supported, including embedded versions. The SDK-based set of programming interfaces provides native integration into most applications.
Thanks to thread safety and efficient resources management, Smart ID Engine SDK allows you to implement scalable AI platforms for face verification processes. Unique algorithms for automatic face detection in selfies, documents and images (scans, photo and video data) allow you to automatically process heterogeneous streams of input images.
Oman Arab Bank
NEC de Colombia
Please fill out the form to get more information about the products,
pricing and trial SDK for Android, iOS, Linux, Windows.