Pattern Recognition and
Image Processing Group
Institute of Visual Computing and Human-Centered Technology
Welcome to PRIP
Modern imaging sensors, like digital photo and video cameras, LIDARs, or medical devices measure huge amounts of data day-by-day. Pattern Recognition and Image Processing aims at the extraction of information from such data.
Simple tasks like the detection of human faces are already solved and solutions are commercially available in digital cameras. More challenging applications such as the image-based autonomous navigation or the detection of anomalies in medical images require systems to incorporate some knowledge to enhance the recognition results and to give semantically appropriate interpretations.
Traditionally, such knowledge could for instance include the characteristics of the camera system or the available processing strategies. More importantly, however, recognition results can be improved by incorporating the composition and structure of the environment. Indeed, the information to be extracted from images is closely related to pieces of data from the same environment that gives them meaning.
Unfortunately, it turns out that methods of statistical pattern recognition cannot cope with the image structure well. The methods of deep-learning too are not yet capable of exploiting the structural nature of images.
At PRIP we are focusing on advanced image representations and methods that allow the structure of the image to become an essential part of recognition systems.
Publication in Scientific Reports receiving attention.
Deep learning for early detection of pathological changes in X-ray bone microstructures: case of osteoarthritis", co-authored by Jiří Hladůvka and published on 27 January 2021 in Scientific Reports (Nature) is receiving high attention.
According to Altmetric it ranks:
- #1 of 57 Scientific Reports articles of similar age (as of April 8th, 2021)
- in the top 5% of all research outputs ever tracked by Altmetric
Best Projects from course Einführung in die digitale Bildverarbeitung (EDBV) WS 2020.
IEEE WIE Best Paper Award @ ACVRW'20.
October 2020: Supervised by Jiří Hladůvka, Verena Renner won the IEEE WIE Best Paper Award @ ACVRW'20 for paper entitled "Towards Identification of Incorrectly Segmented OCT Scans". Congratulations!
Water's gateway to heaven.and with UniWien
Conferences open the doors to show/see newest research
Are you interested in pattern recognition and image processing? Have you ever wondered how it would be like to visit a scientific conference and learn about the state of the art from the researchers in person?
If yes, PRIP-Club supports your active participation. Send your application to Omar Ismail, oismail(at)prip.tuwien.ac.at.
For more information contact us via Email.