Pattern Recognition and
Image Processing Group
Institute of Computer Graphics and Algorithms
PRIP Fakultät für Informatik, TU Wien Technische Universität Wien Fakultät für Informatik

Current Projects


Duration: 2015/01/10 - still running
Description: The aim of this project is to study the composition of image warping algorithms in order to devise a parallel solution involving image pyramids. For panorama stitching of images from multiple cameras, the images need to be warped in order to compensate for differences in perspective and to guarantee smooth transitions. We focus on decentralizing this warping process, so that it can be done inside each camera individually before assembling all images for the final panorama.

Contact: Walter G. Kropatsch, Ines Janusch
Researcher: Hanna Huber, Majid Banaeyan
Partner: Raphael Barth


Duration: 2015/03/24 - still running
Description: The marine annelid Platynereis dumerilii reproduces by external fertilisation and is semelparous, meaning that it spawns only once in its lifecycle. To maximise reproductive success these worms synchronise their spawning events with the lunar cycle and spawn primarily during new moon (Zantke et al., Cell Reports, 2013). In addition, P. dumerilii exhibit particular swimming behaviours during spawning, which are collectively described as a ‘nuptial dance’. The nuptial dance is initiated by excreted pheremones, stimulating male and female worms to swim in circles around one another and finally release sperm and eggs into the water. The aim is to develop methods that enable 2D tracking of spawning worms from captured video and extract data to quantify specified behaviours.

Contact: Walter G. Kropatsch, Nicole M. Artner
Researcher: Daniel Pucher
Partner: Stephanie Bannister

LBP pyramids

Duration: 2014/06/01 - still running
Description: In this project we show how topological categories can be used to identify and remove redundant structural information in a graph representation of an image. Additionally we present two concrete applications, namely an approach to obtain a minimal structural representation of an image, and an image segmentation algorithm that preserves regions in a hierarchy based on the amount of texture.

Contact: Walter G. Kropatsch, Ines Janusch
Researcher: Martin Cerman
Partner: Rocio Gonzales Diaz


Duration: 2015/01/10 - still running
Description: The Spanish Riding School with its Lipizzan horses is one of the most famous tourist attractions in Vienna. In breeding, it is desirable to objectively evaluate the traits of a Lipizzan horse based on quantitative measurements. Unfortunately, traditional approaches of measurement lead to a low repeatability rate due to pose changes of the horse and/or differences between the measurement techniques of different people. The aim is to develop a representation of a 2D horse shape, which is independent of pose OR to develop an approach which is able to “normalize” the current pose of a horse to a standard-pose.

Contact: Walter G. Kropatsch, Nicole M. Artner
Partner: Thomas Druml, Gottfried Brem


Duration: 2012/01/01 - still running
Description: Facial nerve paralysis is a paralysis of the muscles which are innervated by the seventh cranial nerve, resulting in partial or total paralysis of the muscle tone. This causes restrictions of the nerval actuation of muscles responsible for facial expressions which causes asymmetric facial movement. A way to treat facial palsy is to apply neuromuscular reconstruction methods to reestablish the muscle tone and symmetric facial movement.In order to measure the progress quantitatively, physicians require clinical measures extracted from those locations of the face which provide most information about the facial expression. Small artifical markers indicated these locations. These markers are placed manuallly on to the patient's face before the evaluation session begins. During the evaluation session the patient is recorded while performing various facial expressions. In order to evaluate the common condition the distance between two marker positions ist measured. The software which performs the distance calculation needs a manual annotation of the marker points. This task is currently performed by the physician and can take several hours. Object tracking refers to the estimation of the position of objects from an image sequence. Natural objects, such as the human face, have a high potential for deformation and are characterized by an irregular texture. As not only one, but multiple markers have to be tracked simultaneously, additional difficulty is imposed by ensuring that markers that can be uniquely identified.
The aim of this project is to automate the tracking process and to establish markerless tracking of facial features. The markers in the mouth regions could be replaced by methods of visual speech recognition. An alternative to the markers around the eye region is the application of methods used in drowsy driver detection In order to increase the accuracy of the tracking, a reference frame will be denoted manually and then all other frames similar to the reference frame will be marked as anchor frames. Due to the similarity, the tracker can compute the ow between two anchor frames independently. This approach prevents error accumulation in lengthy performances.

Contact: Walter G. Kropatsch, Nicole M. Artner
Researcher: Barbara Koneczny
Partner: Chieh-Han Tzou

Contact: Mail: webmaster(at) | Tel: +43.1.58801.18661 | Fax : +43.1.58801.18697
2014 PRIP, Impressum
This page is maintained by Webmaster ( webmaster(at) ) and was last modified on 13. June 2016 10:52