Head of the Research Group: Dr. András HORVÁTH
Members of the Research Group: Dóra Eszter BABICZ, Bálint MAGYAR, Zsófia MOLNÁR, Adrienn RÁCZ
Contact: horvath.andras@itk.ppke.hu
The Smart Sensory Computing Lab focuses on research in machine vision and artificial intelligence, with a strong emphasis on real-world applications. Our main applied research areas include medical image analysis (e.g., detecting cancer cells in microscopic images and supporting ophthalmic diagnosis based on fundus images), face recognition-based access control systems, and intelligent urban vision systems that detect and track vehicles and pedestrians. Our fundamental research aims to achieve a deeper understanding of learning and vision processes, with a particular focus on improving the generalization capabilities of neural networks . We draw inspiration from the functioning of the human nervous system to develop models that perform well even with limited data.
Our goal is to create intelligent systems that operate reliably with minimal data usage and are capable of making accurate inferences from only a few examples.
Example output of our developed system, which creates a three-dimensional model from two-dimensional ultrasound images and estimates key parameters such as the heart’s ejection fraction.
Future research directions, collaboration opportunities
Our research group offers professional and research collaboration opportunities in the fields of computer vision and artificial intelligence. We have previously participated in several successful grant projects, including OTKA, GINOP, ONR, and EU Horizon programs.
Key publications
- Szabó, G., Bonaiuti, P., Ciliberto, A., & Horváth, A. (2025). Enhancing yeast cell tracking with a time-symmetric deep learning approach. NPJ SYSTEMS BIOLOGY AND APPLI-CATIONS, 11(1), 25.
- Tokodi, M., Magyar, B., Soós, A., Takeuchi, M., Tolvaj, M., Lakatos, B. K., ... & Kovács, A. (2023). Deep learning-based prediction of right ventricular ejection fraction using 2D echocardiograms. CARDIOVASCULAR IMAGING, 16(8), 1005-1018.
- Barna, L., Dudok, B., Miczán, V., Horváth, A., László, Z. I., & Katona, I. (2016). Correlated confocal and super-resolution imaging by VividSTORM. NATURE PROTOCOLS, 11(1), 163-183.
