[IORIU] - [de] - [Object Recognition and Image Understanding]

Object Recognition and Image Understanding [2021 SoSe]
Code
IORIU		 Name
Object Recognition and Image Understanding
LP
8 LP		 Dauer
one semester		 Angebotsturnus
at least every 4th semester
Format
Lecture 4 h + Exercise course 2 h		 Arbeitsaufwand
240h; thereof
90h lectures and tutorials
100h lecture wrap-up and homework
50h preparation for project work, examination		 Verwendbarkeit
B.Sc. Angewandte Informatik,
M.Sc. Angewandte Informatik
M.Sc. Scientific Computing
Sprache
 Lehrende
 Prüfungsschema
Lernziele To have reached understanding of the state-of-the-art in mid- and high-level Computer Vision and to have the ability to relate and contrast different concepts.
To be able to apply essential algorithms from pattern recognition and deep learning to current problems in machine vision.
To be capable of understanding and analyzing the latest publications in Computer Vision and to evaluate their strengths and weaknesses.
To have a firm command of the algorithmic basics and to be able to analyze and solve object recognition problems in novel application areas.
To know the most relevant methods for robust object representation and to judge them based on their applicability and restrictions.
Lerninhalte Methods in mid- and high-level Computer Vision, esp.:
- object detection and classification
- deep learning in Computer Vision and convolutional neural networks
- machine learning approaches for object representation
- video analysis
- recognition of human actions
- local and global feature extraction
- model based approaches
- view based approaches
- generative/discriminative methods
- supervised/unsupervised methods
- registration
- shape analysis
- voting and hashing methods
- hierarchical object representations
Teilnahme-
voraus-
setzungen		 recommended are: solid programming skills and knowledge of basic calculus, statistics, and linear algebra
Vergabe der LP und Modulendnote Bestehen der Modulprüfung
Nützliche Literatur Will be presented in class