Computer Vision 1 (Winter Term 2025/2026)

Overview

• Course consisting of:
  • Lectures (2 hours/week) in BEY/0E40/H (George-Bähr-Str. 1) on Mondays, 11:10–12:40
  • Exercise group sessions (2 hours/week)
    • In BAR/0218/U on Tuesdays, 13:00–14:30
    • In APB/E065/P on Tuesdays, 14:50–16:20
    • In APB/E067/P on Thursdays, 11:10–12:40
    • In APB/E067/P on Thursdays, 13:00–14:30
  • Self-study
  • Final examination
• Lecturer: Bjoern Andres
• Teaching Assistant: Jannik Presberger
• Enrolment (OPAL). Additional rules for enrolment may apply, depending on the study programme.

Contents

• Introduction
• Digital images
  • Color spaces
  • Extrapolation of digital images
  • Interpolation of digital images
• Operators on digital images
  • Point operators
  • Linear operators (esp. convolution)
  • Non-linear operators, including edge and corner detection
• Classification of digital images
  • Logistic regression
  • Gradient descent algorithm
  • Deep learning for computer vision (basics)
  • Backward propagation algorithm
• Segmentation of digital images
  • Image segmentation as a (lifted) multicut problem
  • Semantic segmentation as a node labeling (lifted) multicut problem
  • Local search algorithms
• Object recognition in digital images
  • Single object recognition as a partial quadratic assignment problem
  • Multiple object recognition as a node labeling multicut problem
• Object tracking in digital images
  • Single object tracking as coupled partial quadratic assignment problems
  • Multiple object tracking as an integer linear program
• Linear and integer optimization for computer vision
  • Introduction
  • Simplex algorithm
  • Branch-and-bound/branch-and-cut algorithms