Computer Vision 1 (Winter Term 2022/2023)
Overview
- No lecture on 2023-01-16
- Course (2/2/0) consisting of:
- Lecturer: Bjoern Andres
- Teaching Assistant: Holger Heidrich
- Enrolment (OPAL). Additional rules for enrolment may apply, depending on the study programme.
- Creditable toward the modules CMS-CLS-ELG, CMS-VC-ELG, CMS-VC-ELV1, CMS-VC-ELV2, INF-B-510, INF-B-520, INF-BAS2, INF-BAS7, INF-LE-MA, INF-VERT2, INF-VERT7, MATH-MA-INFGDV
Contents
-
Lectures
- Introduction (slides)
-
Formation of digital images
- Real projective geometry (slides)
- Projective camera (slides)
- Color spaces
-
Operators on digital images
-
Point operators (slides)
- Gamma correction
- Histogram equilibration
- Tone mapping
-
Linear operators (slides)
- Convolution
- Averaging filters
- Derivative filters
-
Non-linear operators (slides)
- Median filter
- Bilateral filter
- Morphological filters
- Component labeling
- Distance transform
- Non-maximum suppression
-
Edge and corner detection (slides)
- Canny edge detector
- Structure tensor
- Harris corner detector
-
Classification of digital images
- Logistic regression (slides)
- Naïve pixel classification (slides)
- Smooth pixel classification as a minimum st-cut problem (slides)
-
Decomposition of digital images
- Seeded region growing heuristics and their limitations
- Multicut and lifted multicut problem
- Local search algorithms
-
Semantic segmentation of digital images (slides)
- Joint graph decomposition and node labeling problem
- Local search algorithm
-
Object recognition in digital images (slides)
- Joint graph decomposition and node labeling problem
- Local search algorithm
-
Object tracking in digital images
- Single object tracking
- Optimization problem
- Solution by dynamic programming
- Multiple object tracking
-
Geometric analysis of digital images
- Panorama stitching
- Optimization problem
- Solution by singular value decomposition
- Stereo vision
- Epipolar geometry
- Optimization problem
- Local search algorithm (RANSAC)