למידה עם רשתות נוירונים

This course introduces the basic concepts and tools for image understanding using Deep Learning. The availability of endless visual data together with modern hardware for parallel image processing, made the Deep Learning architecture a successful framework for visual data analysis with a huge impact in real-world applications. We will start by learning the general architectures of Neural Networks (NN), their underlying concepts, and how we can optimize their parameters using training examples. We then concentrate on Convolutional NN architectures and other recently proposed networks that have been proven to be successful in image processing and image synthesis. We will then study the implementation of the learned concepts for common Computer Vision tasks, such as object detection, object localization, scene segmentation and novel image synthesis using generative adversarial models.