ETH Zurich - D-INFK - IVC - CVG - Lectures - Computer Vision Lab

Computer Vision Lab



The course will start with an info event where the organization of the course will be presented in details. All the students interested in this course should therefore attend the info event.

NEWS: First tutorial: THURSDAY, SEPTEMBER 26, 10:15h (sharp), CNB G 110.


Course Description

The goal of this course is to learn to develop a computer vision system and to gain hands-on-experience. This year's course is focused on mobile computer vision on Android devices. Small teams of students (preferably two) will work on a common project throughout the semester in order to develop a computer vision application. Special attention will be put on technical aspects of computer vision such as 3D reconstruction, camera calibration, feature detection, tracking, virtual reality and their implementation on mobile devices.
While previous experience with Android is not required, at least basic experience with computer vision is mandatory. An introductory tutorial will facilitate the first steps into Android programming.

Lecturers Amaël Delaunoy
Kalin Kolev
Assistant Lorenz Meier
Language English
Location CNB G 110
Credits 10 KP

Prerequisite classes

  • Good programming skills
  • Mandatory: Visual Computing or equivalent (introduction to computer vision required).
    The lecture has to be completed (not in parallel to the course).
  • Advised: Computer Vision, 3D Photography, Computer Graphics or similar classes.

Course Outline

During this course students need to participate in tutorial sessions. The main focus is to carry out a project. At the beginning of the course one or two tutorials will be held. Each student will get assigned one topic (and a teammate) and is expected to work constantly over the whole semester towards the goal of his project. The work should be conducted at home but the experimental lab in CNB D 102.1 could also be used if needed.

Grading will be based on the mid-term and final presentations. The mid-term presentation will account for 25% of the final grade, while 75% will be attributed to the final presentation (including implementation, demo and written report).

Schedule

September 20, 2013Information Event and Introduction
September 26, 2013Android SDK/NDK, OpenCV Basics (SLIDE DOWNLOAD)
October 3, 2013Working plan due
November 14, 2013Mid-Term Presentation
December 19, 2013Final Presentation

Templates

  • Working plan (by October 4, 2013) [doc] [odt]

Student Projects

We encourage students to propose their own ideas, and projects should focus on computer vision on mobile devices. Here are possible directions, more details will be given in the class:

  • Augmented Reality
  • 3D Reconstruction
  • Face Tracking/Reconstruction
  • Computational Photography
  • Kinect on Mobile Phones
  • Image-based Localization

Links


© CVG, ETH Zürich lm@inf.ethz.ch