Diplomarbeiten

Deep Learning-based Forensic Writer Retrieval/Document Segmentation

Status: available Supervisor: Manuel Keglevic Why? Law enforcement agencies possess an extensive collection of handwritten documents. This includes for example documents belonging to open cases, for example death threats, and reference samples from suspects and prisoners. Yet, these collections of documents can only be utilized to a limited extend, since for an identification of an … Continue reading Deep Learning-based Forensic Writer Retrieval/Document Segmentation

Deep Learning-based Image Compression

Status: available Supervisors: Manuel Keglevic, Matthias Wödlinger Why? Traditional lossy image compression techniques like JPEG are pretty old and are honestly only used today because they are built-in in every device that can process images from your camera to your smart watch (and definitely your smart toaster). Since Deep Learning-based end-to-end solutions work pretty well … Continue reading Deep Learning-based Image Compression

Allgemeine Hinweise zu Masterarbeiten

Ablauf Einarbeitung in das Thema und Anmeldung der Arbeit beim Dekanat Implementierung/Ausarbeitung Verfassung der Diplomarbeit Abgabe der fertigen Diplomarbeit beim Dekanat Abschlußpräsentation am Institut (ca. eine Woche vor Diplomprüfung) Diplomprüfung Organisation Auswahl eines Diplomarbeitsthemas aus der Themenliste Zuständigen Betreuer kontaktieren Persönliche Anmeldung bei Ihrem Betreuer Besuch des Seminars während des Semesters In Absprache mit dem … Continue reading Allgemeine Hinweise zu Masterarbeiten

Behaviour detection in a bathroom environment to assist people with dementia

Problem Statement Nursing staff members spend a lot of time on routine tasks, such as assisting in basic activities of daily living (toileting, bathing, transferring, grooming, etc.), ensuring the safety of patients, or walking tours and providing nursing care to their patients. The visit of a toilet is part of the daily routine, and takes … Continue reading Behaviour detection in a bathroom environment to assist people with dementia

Relation Detection: Image Similarity Measurement in Large Historical Archives

Status: available Supervisor: Daniel Helm, Martin Kampel Start: as soon as possible Problem Statement Automatic scene analysis is a crucial task for historians or film archivists in order to preserve and interpret the memories of human cultural history. Film archives include thousands of hours of digitized analog footage from the last 100 years up to … Continue reading Relation Detection: Image Similarity Measurement in Large Historical Archives

Historical Image Restoration: Deep learning-based versus traditional-based deformation strategies

Status: available Supervisor: Daniel Helm, Martin Kampel Start: as soon as possible Problem Statement Automatic video analysis in large historical film collections is a challenging task due to different aspects. One major challenge is related to the quality of historical films, e.g. over-, underexposures and blur. Furthermore, they include scratches, cracks or press cuts and … Continue reading Historical Image Restoration: Deep learning-based versus traditional-based deformation strategies

Exploiting Context in Remote Sensing Object Detection

Status: open Supervisor: Sebastian Zambanini Problem Statement Detecting and classifying objects in remote sensing images is a challenging task, since the objects’ optical signature is less discriminative  due to the small resolution. Common object detection methods in this area aim to find object locations by classifying local structures, while neglecting the larger surroundings of the … Continue reading Exploiting Context in Remote Sensing Object Detection

3D Reconstruction of Ancient Bronze Mirrors

Contact: Simon Brenner With over 3000 known examples, Etruscan mirrors form one of the largest object groups in Etruscan art. The majority are metal (mostly bronze) discs with handles whose polished obverse served as a mirror while the reverse was often embellished with engraved and/or chased figures. The old-fashioned way of documenting these objects is … Continue reading 3D Reconstruction of Ancient Bronze Mirrors

Work place ergonomics: pose estimation by 3D sensing

Problem statement Work place ergonomics is the key factor of a healthy work place. Paying attention to ergonomic standards cannot only prevent back pain or postural deformity, but also vision problems and tensions. The existing 3D sensor fearless detects and tracks the position of an employee during the work and offers possibilities for improving the … Continue reading Work place ergonomics: pose estimation by 3D sensing

Bee data recording from a bee hive in Banja Luka

Problem Statement & Goal We developed a prototype for recording bees when entering the bee hive. The images are used to analyse bee images in various ways: mite detection, pollen quantity estimation, bee counting, … In this practicum we want to record new bee data in bee hives from our partner in Banja Luka. The … Continue reading Bee data recording from a bee hive in Banja Luka