IMAGE ANALYSIS AND COMPUTER VISION

Introduction

On this page, you will find all relevant information regarding your Image Analysis and Computer Vision course, taught by Prof. Luc Van Gool, Prof. Ender Konukoglu, and Prof. Fisher Yu.

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.

Objective

Overview of the most important concepts of image formation, perception and analysis, and Computer Vision. Gaining own experience through theoretical and programming exercises.

Prerequisites

Basic concepts of mathematical analysis and linear algebra. The programming part of the exercises is based on Python and Linux. The course language is English.

Performance Assessment

Different parts of the lecture will be assessed in a maximum 2 hours written exam in English. Doctoral students who participate at the course to earn ECTS points will receive a “Testat” without taking the written examination if their department rules allow this and provided they successfully complete the three lab exercises (interim oral examination). All other students must take the written examination.

Lectures

The lectures will proceed with live attendence every Thursday, between 14:00-17:00, at HG F1. Additionally, the lectures will be live streamed at https://video.ethz.ch/live/lectures/zentrum/hg/hg-f-1.html. The recorded lecture videos will also be made available for later vieweing at https://video.ethz.ch/lectures/d-itet/2021/autumn/227-0447-00L.html.

Exercises

The course includes three lab exercises. The exercises are mandatory only for students who want to receive a "Testat". The exercise sessions will take place every thursday at ETZ D61 from 17:15 to 19:00. The TAs will be available during this time to answer questions. Please check the lab instruction slides for more details about the lab exercises.

Testat guidelines

If you are a doctoral student who want to receive a “Testat”, you need to register for the same before 01/10/2021. You can register by sending a mail with subject “[IACV2021] Testat registration” to goutam.bhat@vision.ee.ethz.ch. The testat students must give a short demo of your solution to the lab exercises to one of the TAs before the corresponding lab deadline. All the exercises have theoretical and programming parts. You will discuss the theoretical parts verbally with the assistants and you will be asked to give a short demo of your working solutions to the programming parts. You need to provide a satisfactory answer to each theoretical question AND demonstrate that your programs work as expected in order for your solution to be deemed complete. To make sure that the presentation process runs as efficiently as possible, only attempt to present once you have everything in working order. Please check the slides for more details about the registration and grading of the exercises for Testat students.

Script

The script for the course is available on Moodle.

Lecture Schedule

Here is a rough lecture schedule for the semester (may be updated during the course of the semester):

Week #

Date

Title

1

23.09.2021

Introduction

2

30.09.2021

Digital Image Formation

3

07.10.2021

Sampling and Enhancement

4

14.10.2021

Feature Extraction and PCA

5

21.10.2021

Color and Texture

6

28.10.2021

Traditional Object Recognition and Segmentation

7

04.11.2021

Optical Flow

8

11.11.2021

3D

9

18.11.2021

3D + Traditional Tracking

10

25.11.2021

Deep Learning - Basics

11

02.12.2021

Deep Learning - CNN

12

09.12.2021

Deep Learning - Supervised I

13

16.12.2021

Deep Learning - Supervised II

14

23.12.2021

Deep Learning - Unsupervised

Exercises

In the first exercise session on 23.10.2021, you will go through an introductory programming task in Python called Exercise 0. Thereafter, you will be provided with the regular exercises handouts. The handout of each exercise will be posted after the deadline of the previous exercise has passed. The solution code and a solution sheet for each exercise will be posted after its deadline has passed.

#

Exercise Sheet and Required Material

Deadline

Solution

0

Introduction to Python [Slides]

None

Practical

1

Basic Image Processing [Slides]

28.10.2021

Theoretical+Practical

2

Stereo Vision

18.11.2021

Theoretical+Practical

3

Image Classification

23.12.2021

Theoretical+Practical

Working Remotely

Please check the lab instruction slides for instructions about working remotely or setting up the lab environment on your own laptop.

Exam instructions

As to the material that has to be studied for the exam of HS21, you should use the slide decks made available to you for this year's lectures as the reference. The material in there has to be studied. The course script can be used in as far as you find it useful to further explain what is in the slides. Material in the course script that is not covered by the slides does not have to be studied.

Sample exam questions from HS19 exam are available on Moodle.



Contract All | Expand All