Lubos Omelina
Post doc
Human Identification

Human identification, particularly face recognition, is a task performed by humans for thousands of years with a very high accuracy. Automatic identification, performed by computers, emerged first from law enforcement applications, but quickly found its way to other civilian applications and the everyday life. As the smart home concept begins to take shape, alongside accessible biometric systems more broadly, an integrated and reliable identification may transform our home to a stream of personalized data. Vast improvements in processing the data can radically change different aspects of our life from comfort to our indispensable health. Modern sensors, e.g. high resolution and depth sensing cameras, became widely used, mainly thanks to the gaming industry. These sensors can provide posture information next to the color image for identifying and tracking multiple people in a home environment. Since the sensors are already deployed, new applications are emerging for instance in gaming or health, like exergames, serious games and physical rehabilitation.

Publications

[1] Dhont, J., Cusumano, D., Boldrini, L., Chiloiro, G., Azario, L., Cellini, F., De Spirito, M., Omelina, L., Vandemeulebroucke, J., Verellen, D. & Valentini, V. Real-time long-term multi-object tracking on cineMR using a tracking-learning-detection framework. Radiotherapy and Oncology, 127:S99-S100, Elsevier Ireland Ltd, 2018.
[2] Bonnechère, .J.P., Omelina, L., Kostkova, K., Van Sint Jan, S. & Jansen, B. The end of active video games and the consequences for rehabilitation. Physiotherapy Research International., 23(4), John Wiley and Sons Ltd, 2018.
[3] Jansen, B., Bonnechère, .J.P., Sholukha, V., Omelina, L. & Van Sint Jan, S. 3D Analysis of Upper Limbs Motion during Rehabilitation Exercises Using the KinectTM Sensor: Development, Laboratory Validation and Clinical Application. Sensors, 18(7), Multidisciplinary Digital Publishing Institute (MDPI), 2018.