Status: available Supervisor: Sebastian Zambanini Problem Statement Learning local image descriptors by means of deep convolutional neural nets [1,2] has recently shown to produce stronger features than traditional hand-crafted ones such as SIFT [3]. However, these nets have been trained and evaluated on general scenarios of (wide-basline) object matching. For the DeVisOr project, matching historical … Continue reading Image Descriptor Learning for Matching Historical Aerial Images with Present-Day Satellite Images →
Status: vergeben Supervisors: Rainer Planinc, Sebastian Zambanini Problemstellung Tiefendaten vereinfachen das Problem des Person Trackings in Videos durch eine deutlichere Unterscheidbarkeit von Vorder- und Hintergund der Szene. In den letzten Jahren wurden zahlreiche Methoden vorgestellt, die aber alle für bestimmte Szenarien gedacht sind und auch dahingehend evaluiert wurden. Es gibt jedoch eine Vielzahl von Parametern, … Continue reading Evaluierung von Person Tracking Algorithmen auf Tiefendaten →
Status: open Supervisors: Robert Sablatnig, Peter Wild Problem Statement See description.
Course description This lecture covers Deep Learning for automatic image and video analysis, such as classifying images into categories or detecting and distinguishing persons. Deep Learning has lead to breakthroughs in these fields; in certain problems, the performance of current methods based on this technology is similar or even better than that of humans – … Continue reading Deep Learning for Visual Computing →
Introduction: FR. 08.04.2016, 13:00 – 15:00, Seminarraum 183/2 Date: FR. 13:00 – 15:00 (see schedule for dates) Place: Seminarraum 183/2 The goal of this lecture/exercise is to get a fundamental understanding of registration techniques used in computer vision, medical imaging, biological imaging and brain mapping. Registration is necessary in order to be able to compare or integrate … Continue reading 2D and 3D Image Registration (VU) →
Status des Praktikums: offen (geplanter Zeitpunkt: März 2016!) Betreuer: Martin Kampel Problemstellung Die Milbe Varroa destructor ist ein Parasit der Honigbiene, saugt Hamolymphe („Bienenblut”) und gilt als Hauptursache für das weltweit auftretende Phänomen des Bienensterbens. Sie ist vollständig an ein Leben im Bienenstock angepasst, vermehrt sich ausschließlich in Bienenbrut und kann ohne Bienenwirte nicht überleben. … Continue reading Rettet die Bienen – mittels Computer Vision →
Status: open Supervisor: Markus Diem, Florian Kleber, Stefan Fiel Text detection is a challenging task when it comes to natural scene images (think of windows or other architectural patterns which have similar properties as printed text). Moreover, if the degree of freedom is increased and rotated text needs to be detected too, simple line searching … Continue reading Text Detection in Natural Scene Images →
Status: open Supervisor: Markus Diem, Florian Kleber, Stefan Fiel Handwritten Text Recognition (HTR) is an open research topic because of the variety of modern, cursive handwriting. The goal – being able to automatically translate handwritten text into machine readable text – is attractive since HTR would render non-digital born documents accessible. The master thesis will … Continue reading Handwritten Text Recognition →
Status: open Supervisor: Markus Diem nomacs – Image Lounge is a light-weight image viewer developed at CVL. It is written in C++ with Qt and OpenCV and published under GPL. Since two years, nomacs has a plugin system which allows for adding modules dedicated to specific tasks. Task A people search plugin will be implemented … Continue reading Face Recognition nomacs Plugin →
Status: closed Supervisor: Markus Diem nomacs – Image Lounge is a light-weight image viewer developed at CVL. It is written in C++ with Qt and OpenCV and published under GPL. Since two years, nomacs has a plugin system which allows for adding modules dedicated to specific tasks. In the last decade Deep Learning has proven … Continue reading Caffe nomacs Plugin →
