Developed in-house using computer vision and deep learning algorithms, FaceSense runs on a Convolutional Neural Network. Tested on 1,00,000+ test cases with a 92% accuracy and 100% precision.
Involves image processing-based pre-processing, Video Analytics-based liveliness detection and Convolutional Neural Network (CNN) based Face Recognition.
Available on multiple channels (Mobile, web, cloud, license, etc.) to allow verification by agents via desktop and mobile phone
Can be scaled up to handle ~1000 concurrent requests in near real time.
Significant error reduction and processing time and an enhanced customer experience.