Deep Learning for Computer Vision - Lecture 14: Visualizing and Understanding

Lecture 14 discusses mechanisms for visualizing and understanding the internal processing of neural networks, and shows how some of these same techniques can be used to generate novel images. We visualize CNN filters and maximally activating patches, and discuss techniques for extracting saliency maps from CNNs. We show how to invert convolutional networks by generating images with gradient ascent, and how this technique can also be used to synthesize adversarial examples. We see how the same strategy of gradient ascent can be used to generate artistic images, including DeepDream, texture synthesis, and neural style transfer. Slides: http://myumi.ch/QA8Bg _________________________________________________________________________________________________ Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving cars. Core to many of these applications are visual recognition tasks such as image classification and object detection. Recent developments in neural network approaches have greatly advanced the performance of these state-of-the-art visual recognition systems. This course is a deep dive into details of neural-network based deep learning methods for computer vision. During this course, students will learn to implement, train and debug their own neural networks and gain a detailed understanding of cutting-edge research in computer vision. We will cover learning algorithms, neural network architectures, and practical engineering tricks for training and fine-tuning networks for visual recognition tasks. Course Website: http://myumi.ch/Bo9Ng Instructor: Justin Johnson http://myumi.ch/QA8Pg