Face Liveness Detection - A tool to prevent spoofing in face recognition systems.
CRMNet: A deep-learning pipeline capable of spotting fake vs legitimate faces and performing anti-face spoofing in face recognition systems. It is built with the help of Keras, Tensorflow, and OpenCV. A sample dataset is uploaded in the sample_dataset_folder.
The problem of detecting fake faces vs real/legitimate faces is treated as a binary classification task. Basically, given an input image, we’ll train a Convolutional Neural Network capable of distinguishing real faces from fake/spoofed faces. There are 4 main steps involved in the task:
This work was done during my internship at SimpleCRM, Nagpur.