Face spoofing detection web application

Background: Face detection/recognition has been one of the most popular deep learning projects/researches recently. One of its daily application is the face verification feature to perform tasks on our devices (e.g., unlocking the device, signing in to some apps, confirming our payment, etc). However, this method could be prone to spoof attacks, in which the model could be fooled with the facial photograph of its respective user (i.e., using a printed or digital picture of the user, and many others face-spoofing attacks). Therefore, a facial anti-spoofing detection would be worth to develop for tackling this malicious problem.

This is a simple web application (with Node.js) that runs Blazeface model, face detection model built with TensorflowJS, to detect faces in a video stream. The face image(s) will be cropped and parsed to a classification model to decide whether it’s a real or spoof face. Kindly visit the repository for more details.

The MobileNet V2 (Sandler, et al., 2019) is used as the classification model and trained with the face anti-spoofing dataset by Timoshenko, et al. 2019. The details of the model development are available in this repository, and the model training notebook is here.