IUPR research focuses on large scale, practical applications of pattern recognition, image understanding, machine learning, and statistical language modeling. The fundamental techniques we are developing are:
Optical character recognition is an excellent area for studying and evaluating approaches to these problems, as well as a useful application. We are currently building a system called OCRopus, designed for large scale digital library applications. A second application is high quality 3D capture of books and art using low-cost digital cameras as part of project DECAPOD. As part of project SICURA, we are developing object recognition and image database methods and software for security applications. IUPR is funded through industrial contract research and public research grants.
opening hours for students
Monday-Friday 2:00 - 4:30 pm
Please see our Jobs Page. For all job-related mail, please send mail to firstname.lastname@example.org
Exam Dates / Prüfungen
For scheduling exam dates (Prüfungen), please contact email@example.com
Please send mail to firstname.lastname@example.org
Some of the organizations that have sponsored our work:
The goal of this semester's bachelor project (master students welcome) is to build a self-replicating 3D printer. That is, build a printer that is capable of printing the parts for more 3D printers...We are going to build a Rostock-style printer, because they have a simple geometry and can be scaled up easily. Eventually, we'd like to build larger versions of this capable of printing car parts, furniture, and other real-world objects.
Technologies / skills:
Home page: http://project13.iupr.com/
This semester's master seminar will deal with topics in statistical NLP. The course NLPA is not a prerequisite but may be helpful. Possible seminar topics include:
Anwendung von Lernmethoden zur Lösung von Problemen, wie Wahrnehmung und Steuerung, Bildverarbeitung, Multimediale Datenbanken, Data Mining, Informationssicherheit, Agentensysteme und autonome Systeme. Verständnis von grundlegenden Konzepten des maschinellen Lernens, der neuronalen Netze, der Bildverarbeitung, Mustererkennung, Entscheidungstheorie und Bayesschen Methoden, Fourietransformationen. Bayesian Methods, Maximum Likelihood Methods, Nearest Neighbor Classification, Feature Extraction, Neural Networks, Perceptrons, Clustering, Vector Quantization, linear and non-linear Filters, Template Matching, Multi-Agent Systems, Self-Organization.
Home page: http://lw.iupr.com
Knowledge of machine learning methods with a focus on massively parallel systems. Understanding of biological analogs, and the computational capabilities and limitations of neural systems. Understanding of other methods and approaches to natural computation. Ability to apply these techniques to the solution of real-world problems. Ability to test and validate massively parallel classifiers. Ability to implement massively parallel classifier and learning systems using GPU and distributioned computations. Basics of machine learning. Statistical foundations. Neural hardware and brain architecture. Simulations of neural systems. Traditional neural network models. Gradient descent learning. Parallel implementations. Deep learning. Clustering. Independent component analysis. Bayesian networks. Reinforcement learning. Testing and application of neuromorphic systems. Self-assembly, self-organization. Multi-agent systems.
Home page: http://ncso.iupr.com/
Building. 48, Room 459
University of Kaiserslautern