VVZ API is not affiliated with ETH Zurich. Data might be outdated or incorrect. Please view the official ETHZ Vorlesungsverzeichnis for binding information.
Computer Vision II
Bilddatenanalyse und Computer Vision II
Last Updated: 2026-02-05 14:55:38
Objective
Introduction into the basic procedures for the interpretation of image content and object recognition. Demonstrating the current capabilities of computer vision systems through selected applications. Gaining own experience through practical computer and programming exercises.
Content
The second part of the course starts with the the basic structure of the human visual system, laying the foundations for more technical aspects, such as the representation of colour. Then follows a discussion of alternative representations of image content, based on unitary transforms, wavelets, Hough transforms, histograms, geometric resampling, multiscale representations, orientation maps, etc. Next the basic issues of image segmentation are discussed. With segmentation, we try to tell the different entities in an image apart. Typically these are the different objects. Several approaches for this crucial step are outlined. Some are fully automatic, others require some input from the user. The structure of digital images covering basic concepts of topology and distance on the discrete image raster will be investigated. Surface characteristics play an important role in object description. Beside colour, which has been discussed in the first part of the course, texture plays an important role. Several techniques for its description are outlined. As to the object shape, focus will be on viewpoint invariant descriptions. This analysis will encompass purely geometric invariants, extracted from edges, and moment invariants that combine shape and surface reflectance information. Based on the introduced rather high-level features, object recognition is possible. Several approaches are outlined, like model-based and view-based schemes. Emerging, mixed schemes are of particular interest. Finally, a selection of real applications are discussed. They serve as examples of how the different topics in the course can be tied together to build useful vision systems.
Resources
Lecture Notes
Course material Script, computer demonstrations, exercises and problem solutions.
General Information
- Language
- German
- Frequency
- Yearly recurring
Examination
- Type
- session examination
- Mode
- oral 30 minutes
Course Components
| Type | Title | Time & Place | Hours |
|---|---|---|---|
| lecture with exercise | Bilddatenanalyse und Computer Vision II |
|
4 h weekly |