Luc Van Gool: Catalogue data in Autumn Semester 2020 |
| Name | Prof. Dr. Luc Van Gool |
| Field | Computer Vision |
| Address | Institut für Bildverarbeitung ETH Zürich, ETF C 117 Sternwartstrasse 7 8092 Zürich SWITZERLAND |
| Telephone | +41 44 632 65 78 |
| vangool@vision.ee.ethz.ch | |
| Department | Information Technology and Electrical Engineering |
| Relationship | Full Professor |
| Number | Title | ECTS | Hours | Lecturers | |
|---|---|---|---|---|---|
| 227-0085-07L | Projects & Seminars: Deep Learning for Smartphone Apps (DLSA)  Only for Electrical Engineering and Information Technology BSc. The course unit can only be taken once. Repeated enrollment in a later semester is not creditable. | 3 credits | 3P | L. Van Gool | |
| Abstract | The category of "Laboratory Courses, Projects, Seminars" includes courses and laboratories in various formats designed to impart practical knowledge and skills. Moreover, these classes encourage independent experimentation and design, allow for explorative learning and teach the methodology of project work. | ||||
| Objective | Deep Learning with Smartphone Sensors – Programming Android Phones – Neural Networks – Keras/TensorFlow -- Projects on Smartphones. Latest smartphone generations are equipped with computational capabilities (CPU, GPU, NPU, DSP) matching common PCs from a decade ago. Moreover, smartphones have several sensors that can acquire many useful information beyond audio and visual data, for instance where we are, what we are doing, with whom we are together, what is our body constitution, what are our needs. Based on this information our smartphone offers us the appropriate computational power to process them in loco without sending the sensor data to the cloud. This course focuses on giving the bases of machine (deep) learning and embedded systems. Students will learn the tools to implement machine/deep learning algorithms in their Android phones to be smarter. The course will end with a 4 weeks project where the students can target a specific application scenario. The course will be taught in English. | ||||
| 227-0085-11L | Projects & Seminars: Deep Learning for Image Manipulation (DLIM) Only for Electrical Engineering and Information Technology BSc. The course unit can only be taken once. Repeated enrollment in a later semester is not creditable. | 3 credits | 3P | L. Van Gool | |
| Abstract | The category of "Laboratory Courses, Projects, Seminars" includes courses and laboratories in various formats designed to impart practical knowledge and skills. Moreover, these classes encourage independent experimentation and design, allow for explorative learning and teach the methodology of project work. | ||||
| Objective | Deep Learning – Image Manipulation – Image Enhancement – Image Restoration – Style Transfer – Image to Image Translation – Generative Models – TensorFlow/PyTorch – Projects With the advent of deep learning tremendous advances were achieved in numerous areas from computer vision, computer graphics, and image processing. Using these techniques, an image can be automatically manipulated in various ways with high-quality results, often fooling the human observer. Deep learning based image processing and manipulation are being applied in a vast number of emerging technologies, including image enhancement in smartphone cameras, automated image editing, image content creation, graphics, and autonomous driving. This course focuses on the fundamentals of deep learning and image manipulation. Students will learn the tools to implement and develop deep learning solutions for a variety of image manipulation tasks. The course will end with a 4 weeks project where the students can target a specific application scenario. The course will be taught in English. | ||||
| 227-0085-24L | Projects & Seminars: Vision and Control in RoboCup Only for Electrical Engineering and Information Technology BSc. The course unit can only be taken once. Repeated enrollment in a later semester is not creditable. | 1 credit | 1P | L. Van Gool | |
| Abstract | The category of "Laboratory Courses, Projects, Seminars" includes courses and laboratories in various formats designed to impart practical knowledge and skills. Moreover, these classes encourage independent experimentation and design, allow for explorative learning and teach the methodology of project work. | ||||
| Objective | RoboCup is a tournament where teams of autonomous robots compete in soccer matches against each other. The ETH team NomadZ plays in the standard platform league with the humanoid NAO robot, where the focus lies on developing robust and efficient algorithms for vision, control and behavior. In this course, the basic challenges we encounter in RoboCup are presented and approached in practical exercises using MATLAB and Python. The topics cover visual localization, deep learning for object detection and reinforcement learning for control. The course is offered to students of the 5th semester. | ||||
| 227-0085-31L | Projects & Seminars: Vision Goes Vegas Only for Electrical Engineering and Information Technology BSc. The course unit can only be taken once. Repeated enrollment in a later semester is not creditable. | 2 credits | 2P | L. Van Gool | |
| Abstract | The category of "Laboratory Courses, Projects, Seminars" includes courses and laboratories in various formats designed to impart practical knowledge and skills. Moreover, these classes encourage independent experimentation and design, allow for explorative learning and teach the methodology of project work. | ||||
| Objective | Computer Vision beschäftigt sich unter anderem damit, Maschinen zu befähigen ihre Umwelt zu sehen und das wahrgenommene Bild zu verstehen. In unserem Projekt soll ein System entwickelt werden, das Spielkarten erkennen kann und, einer guten Strategie folgend, erfolgreich Black-Jack spielen kann. Die Teilnehmer des Projektes werden kleine Teams bilden und gemeinsam mit einem Assistenten die Aufgabe erarbeiten und eine Implementierung erstellen. Am Ende des Semesters sollen die Programme im öffentlichen Wettstreit gegeneinander antreten! Ziel des Projektes ist es, aktuelle Methoden der Computer Vision kennen zu lernen. Spielkarten, die von einer Digitalkamera in beliebiger Orientierung aufgenommen werden, müssen registriert und erkannt werden. Ein Strategiemodul kontrolliert dann die Spieltaktik aufgrund allgemeiner Regeln und dem Wissen über schon gefallene Karten. Da sehr viele verschiedene Möglichkeiten bestehen, solch ein System zu realisieren, sind der Phantasie der Teilnehmer keine Grenzen gesetzt. Als Voraussetzungen sollte Interesse an Computer Vision mitgebracht werden und die Bereitschaft, sich in einem Team von Mitstudierenden einzubringen. Kenntnisse in C++ sind notwendig. Dieses P&S wird in englischer Sprache durchgeführt. | ||||
| 227-0447-00L | Image Analysis and Computer Vision  | 6 credits | 3V + 1U | L. Van Gool, E. Konukoglu, F. Yu | |
| Abstract | Light and perception. Digital image formation. Image enhancement and feature extraction. Unitary transformations. Color and texture. Image segmentation. Motion extraction and tracking. 3D data extraction. Invariant features. Specific object recognition and object class recognition. Deep learning and Convolutional Neural Networks. | ||||
| Objective | Overview of the most important concepts of image formation, perception and analysis, and Computer Vision. Gaining own experience through practical computer and programming exercises. | ||||
| Content | This course aims at offering a self-contained account of computer vision and its underlying concepts, including the recent use of deep learning. The first part starts with an overview of existing and emerging applications that need computer vision. It shows that the realm of image processing is no longer restricted to the factory floor, but is entering several fields of our daily life. First the interaction of light with matter is considered. The most important hardware components such as cameras and illumination sources are also discussed. The course then turns to image discretization, necessary to process images by computer. The next part describes necessary pre-processing steps, that enhance image quality and/or detect specific features. Linear and non-linear filters are introduced for that purpose. The course will continue by analyzing procedures allowing to extract additional types of basic information from multiple images, with motion and 3D shape as two important examples. Finally, approaches for the recognition of specific objects as well as object classes will be discussed and analyzed. A major part at the end is devoted to deep learning and AI-based approaches to image analysis. Its main focus is on object recognition, but also other examples of image processing using deep neural nets are given. | ||||
| Lecture notes | Course material Script, computer demonstrations, exercises and problem solutions | ||||
| Prerequisites / Notice | Prerequisites: Basic concepts of mathematical analysis and linear algebra. The computer exercises are based on Python and Linux. The course language is English. | ||||
| 227-0919-00L | Knowledge-Based Image Interpretation | 0 credits | 2S | L. Van Gool | |
| Abstract | With the lecture series on special topics of Knowledge based image interpretation we sporadically offer special talks. | ||||
| Objective | To become acquainted with selected, recent results in image analysis and interpretation. | ||||