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

Computer Vision Lab

Lecturers: Kevin Koeser, Kalin Kolev

Assistant: Lorenz Meier


The course will start with an info event where possible projects are suggested. All the students interested in this course should therefore attend the info event.


Course Description

The goal of this course is to learn to develop a computer vision system and to gain hands-on-experience. Important components of a computer vision system, like camera systems, camera interfaces, image processing libraries, camera calibration etc., will be explained tutorial-like first and then used in projects to get in-depth understanding. This year's course is focused on mobile computer vision on Android devices. Each student or small teams of students will work on one project throughout the semester.


Grading will be based on the implementation, the final demo and written report. The final deadline for the course projects is December 21, and a mid-term presentation has to be held on November 9. Projects which have reached 60% of the functionality and show a working demo version on the mid-term presentation get a bonus on their final grade.


There are no formal prerequisites, Computer Vision I or II or 3D Photography are however recommended.

Course Outline

During this course students need to participate in tutorial sessions and carry out a project. At the beginning of the course tutorials will be held. Each student gets assigned one topic and works constantly over the whole semester towards the goal of his project. The work is carried partly at home, partly in the experimental lab in CNB D 102.1.

Tutorial Lectures

The following tutorials will be held. Slides will be available ahead of lectures.

  • Information Event (September 21, 2012) (slides)
  • Android SDK/NDK (September 28, 2012) (slides)
  • OpenCV Basics and Camera Calibration (October 5, 2012) (slides)


  • Working plan (by October 5, 2012) [doc] [odt]

Student Projects

Please feel free to propose your own topics as well, if they fit we will try to accomodate them. Taken topics are marked accordingly.

  • Silhouette-based live 3D reconstruction (live visual hull estimation based on object/background color models, use Qualcomm's Vuforia toolkit to get camera poses)

  • Stefan Cyril Götschi & Enes Poyraz

    suggested literature:
    D. Snow, P. Viola, R. Zabih, "Exact Voxel Occupancy with Graph Cuts" [pdf]
    C. Rother, V. Kolmogorov, A. Blake, "GrabCut — Interactive Foreground Extraction using Iterated Graph Cuts" [pdf]

  • Stereo-based live 3D reconstruction (live depth map estimation and fusion, use Qualcomm's Vuforia toolkit to get camera poses)

  • suggested literature:
    R. Newcombe, A. Davison, "Live Dense Reconstruction with a Single Moving Camera" [pdf]

  • Markerless augmented reality on a cell phone (use classical structure-from-motion approaches to estimate a ground plane and relative camera poses)

  • David Samuelsson & Alexey Sizov

    suggested literature:
    R. Hartley, A. Zisserman, "Multiple View Geometry in Computer Vision" (especially the chapters on homography estimation)

  • 3D reconstruction on a stereo tablet (use a combination of ICP/SfM to get camera poses and depth map fusion for 3D modeling)

  • suggested literature:
    R. Newcombe et al., "KinectFusion: Real-Time Dense Surface Mapping and Tracking" [pdf]

  • On-site event re-living (camera pose estimation based on feature matching and Direct Linear Transform in a client/server platform, full geometry of the scene and a feature data base will be provided)

  • suggested literature:
    L. Ballan, G. Brostow, J. Puwein, M. Pollefeys, "Unstructured Video-Based Rendering: Interactive Exploration of Casually Captured Videos" [pdf]


© CVG, ETH Zürich