Stanford CS231n: Convolutional Neural Networks for Visual Recognition

Stanford's deep learning for computer vision course (Spring 2017), taught by Fei-Fei Li, Justin Johnson and Serena Yeung. Lectures cover image classification, loss functions and optimisation, backpropagation and neural networks, convolutional networks, training neural networks, deep learning software, CNN architectures, recurrent networks, detection and segmentation, visualising networks, generative models and deep reinforcement learning. It suits university students with Python, linear algebra and calculus who want serious deep-learning foundations. 3Blue1Brown's Neural Networks is a good warm-up.