Overview
Learn to create a sophisticated facial recognition application using Python, TensorFlow, and Kivy in this comprehensive tutorial series. Begin by setting up TensorFlow and Keras for deep learning, then progress to collecting image samples and utilizing the Labeled Faces in the Wild dataset. Master loading images into the TensorFlow Dataloader and construct a Siamese Neural Network based on the "Siamese Neural Networks for One-shot Image Recognition" paper. Develop a custom training loop with tf.GradientTape, test the model on various images, and integrate it with OpenCV. Finally, build a Kivy app and incorporate the TensorFlow model for a fully functional facial recognition authentication system. Access provided code repositories, research papers, and datasets to enhance your learning experience throughout this hands-on, step-by-step guide to advanced facial recognition technology.
Syllabus
- Start
- Setting Up Tensorflow and Keras for Deep Learning
- Collecting Image Samples and Using the LFW Dataset
- Loading Images into the Tensorflow Dataloader
- Building a Siamese Neural Network
- Building a Custom Training Loop with tf.GradientTape
- Testing the Model on Images
- Integrating with OpenCV
- Building the Kivy App and Integrating with Tensorflow
Taught by
Nicholas Renotte