Ing. Tomáš Kerepecký, Ph.D.

postdoc

kerepecky@utia.cas.cz 266052864

Department: Department of Image Processing

Research interests: Inverse problems, machine learning, artificial intelligence

Biography

Publication list

Tomáš Kerepecký received the M.Sc. degree in Computational Physics from the Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University (CTU), Prague, in 2017. He began his Ph.D. studies the same year at ELI Beamlines and the Institute of Physics of the Czech Academy of Sciences (CAS), focusing on computational simulations of laser–plasma interactions. His work in this area sparked a growing interest in data analysis and image processing, which motivated him to shift fields. In 2019, he started a Ph.D. in Mathematical Engineering at CTU, in collaboration with the Institute of Information Theory and Automation (UTIA), CAS. From 2021 to 2022, he was a Fulbright visiting scholar at Washington University in St. Louis, gaining international research experience. He completed his Ph.D. in 2025, with research focused on inverse problems in image restoration and artificial intelligence. He is currently a postdoctoral researcher at UTIA, CAS, and is also pursuing an M.A. in Practical Theology and Leadership at TCM International Institute, Austria.

More info: here

Books and chapters

Kerepecký Tomáš, Šroubek Filip, Zitová Barbara, Flusser Jan: Automated Actor Recognition in Video Content, Data Science in Applications : Towards AI-Driven Approaches, p. 3-22, Eds: Dzemyda G., Bernatavičienė J., Kacprzyk J. Download Download DOI: 10.1007/978-3-031-88486-3_1 [2025]

Journal articles

Kerepecký Tomáš, Šroubek Filip, Flusser Jan: Implicit neural representation for image demosaicking, Digital Signal Processing 159 Download Download DOI: 10.1016/j.dsp.2025.105022 [2025]
Karella Tomáš, Suk Tomáš, Košík Václav, Bedratyuk L., Kerepecký Tomáš, Flusser Jan: 3D Non‑separable Moment Invariants and Their Use in Neural Networks, SN Computer Science 5 Download Download DOI: 10.1007/s42979-024-03504-x [2024]

Other publications

Kerepecký Tomáš, Šroubek Filip: Inverse Problems in Image Restoration, Inverse Problems: Modelling and Simulation : Extended Abstracts of the IPMS Conference 2024, p. 107-113, Eds: Hasanoğlu A. H., Novikov R., Van Bockstal K. Download Download DOI: 10.1007/978-3-031-87213-6_14 [2025]
Kerepecký Tomáš, Šroubek Filip, Zitová Barbara, Flusser Jan: STAR: Screen Time and Actor Recognition in Video Content, Pattern Recognition : 46th DAGM German Conference, DAGM GCPR 2024, p. 270-284 Download Download DOI: 10.1007/978-3-031-85187-2_17 [2025]
Kerepecký Tomáš, Šroubek Filip, Novozámský Adam, Flusser Jan: NeRD: Neural field-based Demosaicking, Proceedings of the 2023 IEEE International Conference on Image Processing (ICIP), p. 1735-1739 Download DOI: 10.1109/ICIP49359.2023.10221948 [2023]
Staněk R., Kerepecký Tomáš, Novozámský Adam, Šroubek Filip, Zitová Barbara, Flusser Jan: Real-Time Wheel Detection and Rim Classification in Automotive Production, Proceedings of the 2023 IEEE International Conference on Image Processing (ICIP), p. 1410-1414 Download DOI: 10.1109/ICIP49359.2023.10223161 [2023]
Kerepecký Tomáš, Liu J., Ng X. W., Piston D. W., Kamilov U. S.: Dual-Cycle: Self-Supervised Dual-View Fluorescence Microscopy Image Reconstruction using CycleGAN, Proceedings of the 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE (Rhodes Island, Greece, 2023) Download DOI: 10.1109/ICASSP49357.2023.10095386 [2023]
Kerepecký Tomáš, Šroubek Filip: D3Net: Joint Demosaicking, Deblurring and Deringing, 2021 25th International Conference on Pattern Recognition (ICPR), p. 8283-8290 Download DOI: 10.1109/ICPR48806.2021.9413121 [2021]
Šroubek Filip, Kerepecký Tomáš, Kamenický Jan: Iterative Wiener Filtering for Deconvolution with Ringing Artifact Suppression, Proceedings of the 27th European Signal Processing Conference (EUSIPCO 2019), p. 1159-1163 Download DOI: 10.23919/EUSIPCO.2019.8903114 [2019]