Research

Karel Horak

Artificial Intelligence | Machine Vision Machine learning algorithms in data & image analysis. Object recognition and anomaly detection in industry, biomedical and cultural heritage tasks. Projects & affiliation: Cybernetics and Artificial Intelligence (Czech Institute of Informatics, Robotics and Cybernetics). Machine Learning Methods in Computer Vision (Technische Universität Wien). Deep Learning in High-Performance Vision (Austrian Institute of … Continue reading Karel Horak

VisuAAL

Privacy-Aware and Acceptable Video-Based Technologies and Services for Active and Assisted Living VisuAAL is a transdisciplinary and cross-sectoral European project, addressing the secure use of video-based AAL technologies to deliver effective and supportive care to older adults managing their health and wellbeing. Researchers from the fields of Computer Science, Engineering, Healthcare, Law, Business and Sociology … Continue reading VisuAAL

SatPark

Satellite image based on-street parking availability technology Finding a free parking spot in a city is tiring. It also costs a lot of time, money and nerves, and, last but not least, causes a lot of  unnecessary emission. Top notch parking navigation software systems are still not able to ease these pains, as options to … Continue reading SatPark

WCBuddy

Investigation of behaviour modelling on the toilet for supporting the autonomous use via guidance In the field of toileting (personal hygiene), there are surprisingly few AAL / supporting technologies and also few projects dealing with this topic. This is all the more surprising, as there is a great need and the visit to a toilet … Continue reading WCBuddy

IPT Dataset

Identity Preserved Tracking (IPT) Dataset   Description The IPT (Identity Preserved Tracking) dataset consisting of 10 sequences of depth data recorded using an Orbbec Astra depth sensor. It features sequences in ten different locations with a high amount of background variation and is designed to be applicable to a wide range of tasks. Its labeling is … Continue reading IPT Dataset

MaLeStV

Machine Learning of motion patterns in the penal system The everyday life of a prison is characterised by aspects of security. A number of structural and organisational measures (prison cell design, video surveillance, …) are designed to reduce the risk of safety-related incidents in prisons and to support the personnel of the correctional system in … Continue reading MaLeStV

FairAlgos

Fairness, Bias und Transparenz von Computer Vision Algorithmen Datenbasierte Technologie hat das Ziel, nach gewissen Merkmalen zu diskriminieren, zu klassifizieren und zu unterscheiden. Hierbei muss jedoch zwischen erwünschten und unfairen Formen der Diskriminierung unterschieden werden. In aktuellen und weit verbreiteten Bilddatenbanken wie zum Beispiel dem IMDB Datenset wurde ein geschlechtsspezifischer Bias offengelegt. Das bedeutet, dass … Continue reading FairAlgos

MyeFLOW

Automated minimal residual disease assessment in childhood acute myeloid leukemia Acute leukemia is the most frequent cancer in children and adolescents. Assessment of the treatment response via quantification of minimal residual disease (MRD) at certain time-points of treatment has become a fundamental diagnostic procedure for tailoring therapy intensity and duration. In particular, children with the … Continue reading MyeFLOW

PCB DSLR Dataset

The PCB DSLR dataset is meant to facilitate research on computer-vision-based Printed Circuit Board (PCB) analysis, with a focus on recycling-related applications. The dataset contains 748 images of PCBs from a recycling facility, captured under representative conditions using a professional DSLR camera. All images come with accurate PCB segmentation information as well as bounding box information … Continue reading PCB DSLR Dataset