Coders Packet

Face Detection using MTCNN and Tensorflow in Python

By Rigved Alankar

The project involves face detection to recognize five eminent personalities using MTCNN a pretrained CNN model, to detect face in the image and then training a classifier to classify the images.

Face detection is a computer technology being used in a variety of applications that identifies human faces in digital images. Its application ranges from facial recognition system to smarter advertising in the marketing sector to find missing persons which helps police to solve kidnappings. The project involves face detection to recognize five eminent personalities - Donald Trump, Narendra Modi, Jack Ma, Elon Musk and Bill Gates using MTCNN (Multi-task Cascaded Convolutional Neural Networks) which is an algorithm consisting of 3 stages, which detects the bounding boxes of faces in an image along with their 5 Point Face Landmark. Each stage gradually improves the detection results bypassing its inputs through a CNN, which returns candidate bounding boxes with their scores, followed by non-max suppression. After detecting faces in an image using MTCNN, we then create face embeddings by employing standardization and then we create an SVM classifier to classify images.

 

Download Complete Code

Comments

No comments yet