1. Handbook
  2. Subjects
  3. Computer Vision

Computer Vision (COMP90086)

Graduate courseworkPoints: 12.5On Campus (Parkville)

You’re viewing the 2025 Handbook:
Or view archived Handbooks

Overview

Availability
Semester 2
FeesLook up fees

AIMS

From self-driving cars to automatic processing of medical scans, vision is a key sensory modality for a variety of artificial intelligence tasks. However, extracting meaning from images poses various computational challenges. In this subject, students will learn the basic principles of image formation and computational methods for interpreting images. Students will develop an understanding of the standard frameworks used in computer vision algorithms and their applications in tasks such as object recognition, target detection, and three-dimensional reconstruction. The programming language used is Python.

INDICATIVE CONTENT

Topics covered may include:

  • Basics of image formation
  • Illumination and reflectance models
  • Colour spaces
  • Feature detectors and descriptors
  • Stereo correspondence
  • Methods for recovering three-dimensional shape
  • Image segmentation
  • Categorical and instance-level recognition

Intended learning outcomes

On completion of this subject, students should be able to: 

  • Identify basic challenges involved in interpreting image data 
  • Understand and articulate the mathematical and/or algorithmic basis of common techniques used in computer vision 
  • Analyse, design, and implement relevant techniques in computer vision to given problems 
  • Undertake end-to-end research experiments, including evaluation with image datasets, including reporting results
  • Communicate technical solutions to computer vision problems

Generic skills

  • Ability to undertake problem identification, formulation, and solution
  • Ability to utilise a systems approach to complex problems and to design and operational performance ability to manage information and documentation
  • Capacity for creativity and innovation
  • Ability to communicate effectively with both technical experts and the community at large

Last updated: 8 November 2024