Tomas Svoboda

Tomas Svoboda

Professor
Faculty of Electrical Engineering (vicedean)
Department of Cybernetics
Vision for Robotics and Autonomous Systems (head)
Appointments on request. calendar, not all events are visible.

Czech Technical University in Prague, Fac. of Electrical Engineering, Dep. of Cybernetics
Center for Machine Perception, Center for Robotics and Autonomous Systems
Karlovo namesti 13, 121 35 Prague, Czech Republic
Room1: KN:E-226, tel:+420 2 24357448, fax:+420 2 24357385
e-mail: svobodat AT fel.cvut.cz
LinkedIn: svobodatomas

Hiring

I am often looking for strongly motivated PhD as well as undergraduate students. Occasionally, I also have an opening for post-doc positions. If you are interested in what we are doing (check also the VRAS group pages) and want to join us, send me your resume.

Teaching

A book about (Matlab) programming for Image Processing and Computer Vision. Free codes, demos, links to presentations, etc.
Student projects: Open topics: [CS | EN]. Older. Finished theses (not a complete list, many older works are not there).
B4B33RPH Solving problems and (other) games. Bachelor course in the study program Open Informatics. [maze video]
B3B33KUI Cybernetics and Artificial Intelligence. Introducing students into the search, sequential decisions, reinforcement learning and simple classifiers. Also in English as BE5B33KUI
Open Informatics, a computer science program, member of the board.
Cybernetics and Robotics, bachelor and master program, member of the board.
Electrical Engineering and Computer Science, EECS bachelor program, (2015-2021) head of the board.
Hry, Algoritmy, Programovani, a jine hratKY. (Games, Algorithms, Programming, and gaming ...) for (not only) high-school students.

past teaching

BE5B33PRG Programming Essentials. Introductory programming course for the EECS study program.
A4M33MPV Computer Vision Methods. Master course in the study program Open Informatics.
XE383ZS Signal and Image Processing. X383ZS in Czech. Demo video.
XE33PVR Computer Vision and Virtual Reality.
X33KUI Cybernetics and Artificial Intelligence (in Czech)
Y33ROV Robotics and Machine Perception (in Czech), demo video.

Research Interests, Projects

Appoximately in reverse chronological order. You may also check related departmental demos/news, VRAS YouTube channel.

Resilient machines through continuous learning and sensing, part of project Roboprox-Robotics and advanced industrial production. Vacant positions!
Search and Rescue. Our CTU-CRAS team qualified to DARPA SubT (#subtchallenge). We won the prize at the Tunnel Circuit and at the Urban Circuit.
Research Center for Informatics, RCI.
perception for automotive, H2020 project Enable-S3. Public event Dublin 2018/06 video.
vision for aerial robots, participating on MBZIRC.
Rescue robotics. Vision for outdoor rescue robots (UGV), Eurathlon/TRADR summer school, European (EU) projects TRADR, Nifti, CMP demos, NIFTi robot in Czech TV
Advanced predictors for object detection and tracking.
Learning for tracking and recognition.
Multicamera environments, Virtual Rooms. See also ViRoom, Blue-C.
(Multi-)Camera (self-)calibration. Free SW downloadable
Omnidirectional vision; Omnivis2004, The fifth Workshop on Omnidirectional Vision, Camera Networks and Non-classical cameras, In conjunction with ECCV2004.

Selected publications

P. Vacek, D. Hurych, K. Zimmermann, and T. Svoboda. Let-It-Flow: Simultaneous Optimization of 3D Flow and Object Clustering. IEEE Transactions on Intelligent Vehicles, 2024, 10.1109/TIV.2024.3443316, arXiv, codes, data.
Z. Straka, T. Svoboda, M. Hoffmann. PreCNet: Next-Frame Video Prediction Based on Predictive Coding. IEEE Transactions on Neural Networks and Learning Systems, Vol 35, Issue 8, 2024, 10.1109/TNNLS.2023.3240857, arXiv.
R. Agishev, K. Zimmermann, V. Kubelka, M. Pecka, T. Svoboda. MonoForce: Self-supervised Learning of Physics-aware Model for Predicting Robot-terrain Interaction. To appear at IEEE/RSJ IROS2024, arXiv, codes, data.
Petrlík, M., Petráček, P., Krátký, V., Musil, T., Stasinchuk, Y., Vrba, M., Báča, T., Heřt, D., Pecka, M., Svoboda, T., and Saska, M. (2024). UAVs Beneath the Surface: Cooperative Autonomy for Subterranean Search and Rescue in DARPA SubT. IEEE Transactions on Field Robotics, early access 2024/11, arXiv, 10.1109/TFR.2024.3492160
A. H. Gebrehiwot, D. Hurych, K. Zimmermann, P. Perez, T. Svoboda. T-UDA: Temporal Unsupervised Domain Adaptation in Sequential Point Clouds. IEEE/RSJ IROS2023. arXiv, codes, data.
A. H. Gebrehiwot, P. Vacek, D. Hurych, K. Zimmermann, P. Perez, T. Svoboda. Teachers in Concordance for Pseudo-Labeling of 3D Sequential Data. IEEE Robotics and Automation Letters, February 2023, Vol 8, Issue 2, 10.1109/LRA.2022.3226029. Also presented at ICRA2023. arXiv, codes.
P. Vacek, O. Jasek, K. Zimmermann, T. Svoboda. Learning to Predict Lidar Intensities. IEEE Transactions on Intelligent Transportation Systems. 10.1109/TITS.2020.3037980. April 2022, Volume 23, Issue 4.
V. Salansky, K. Zimmermann, T. Petricek, T. Svoboda. Pose consistency KKT-loss for weakly supervised learning of robot-terrain interaction model. IEEE Robotics and Automation Letters. 10.1109/LRA.2021.3076957. 2021, Volume 6, Issue 3. [KKT-loss web]
V. Kubelka, M. Reinstein, T. Svoboda. Tracked Robot Odometry for Obstacle Traversal in Sensory Deprived Environment. IEEE/ASME Transactions on Mechatronics. Dec 2019, Vol. 24, Issue: 6. 10.1109/TMECH.2019.2945031
Tomáš Petříček, Vojtěch Šalanský, Karel Zimmermann, Tomáš Svoboda. Simultaneous exploration and segmentation for search and rescue. Journal of Field Robotics. 36(4), June 2019, pp 696-709 10.1002/rob.21847, read-only version.
Martin Pecka, Karel Zimmermann, Matěj Petrlík, Tomáš Svoboda. Data-driven Policy Transfer with Imprecise Perception Simulation. IEEE Robotics and Automation Letters, Vol. 3, Issue 4, Oct 2018, also to be presented at IROS2018. [submitted PDF@arXiv, IEEE Xplore]
K. Zimmermann, T. Petricek, V. Salansky, T. Svoboda. Learning for Active 3D Mapping. In International Conference on Computer Vision (ICCV) 2017. Accepted as Oral presentation(!). ICCV PDF arXiv PDF, DemoVideo, ICCV talk
M. Pecka, K. Zimmermann, T. Svoboda. Fast Simulation of Vehicles with Non-deformable Tracks. In Intelligent Robots and Systems (IROS), 2017. arXiv PDF
M. Pecka, V. Salansky, K. Zimmermann, T. Svoboda. Autonomous flipper control with safety constraints. In Intelligent Robots and Systems (IROS), 2016. IEEE Xplore; pdf.
V. Kubelka, M. Reinstein, and T. Svoboda. Improving Multimodal Data Fusion for Mobile Robots by Trajectory Smoothing. In Robotics and Autonomous Systems. Vol 84, pp 88-96, Oct 2016. accepted pdf. at publisher site. j.robot.2016.07.006.
M. Pecka, K. Zimmermann, M. Reinstein, and T. Svoboda. Controlling Robot Morphology from Incomplete Measurements. In IEEE Transactions on Industrial Electronics, Feb 2017, Vol 64, Issue: 2, pp. 1773-1782 [pdf (before proof read), IEEE Xplore]
V. Kubelka, L. Oswald, F. Pomerleau, F. Colas, T. Svoboda, and M. Reinstein. Robust data fusion of multi-modal sensory information for mobile robots. In Journal of Field Robotics, June 2015, Vol 32, Issue 4. [pdf]
Martin Pecka and Tomas Svoboda. Safe Exploration Techniques for Learning Reinforcement An Overview. LNCS 8906, 2014, PDF.
K. Zimmermann, D. Hurych, T. Svoboda. Non-Rigid Object Detection with Local Interleaved Sequential Alignment (LISA). In IEEE Transactions on Pattern Analysis and Machine Intelligence, [pdf as accepted], April 2014, Vol 36, Issue 4
K. Zimmermann, D. Hurych, T. Svoboda. Exploiting Features -- Locally Interleaved Sequential Alignment for Object Detection. In 11th Asian Conference on Computer Vision - ACCV 2012, [pdf]
T. Petricek, T. Svoboda. Area-weighted Surface Normals for 3D Object Recognition. In 21st International Conference on Pattern Recognition - ICPR 2012 [pdf]
K. Zimmermann, T. Svoboda, J. Matas. Anytime learning for the NoSLLiP tracker. Image and Vision Computing. 27(11), 2009, [pdf].
K. Zimmermann, J. Matas, and T. Svoboda. Tracking by an Optimal Sequence of Linear Predictors. IEEE Transactions on Pattern Analysis and Machine Intelligence. 31(4), 2009, [pdf]
K. Zimmermann, T. Svoboda, J. Matas. Simultaneous learning of motion and appearance. The 1st International Workshop on Machine Learning for Vision-based Motion Analysis, In conjunction with the 10th European Conference on Computer Vision 2008, [ paper | talk ].
K. Zimmernann, T. Svoboda, J. Matas. Adaptive Parameter Optimization for Real-time Tracking. In Proceedings of the Workshop on Non-rigid Registration and Tracking through Learning - NRTL 2007 (joint workshop with the ICCV 2007). [ paper ]
J. Matas, K.Zimmermann, T. Svoboda, A. Hilton. Learning Efficient Linear Predictors for Motion Estimation. In Proceedings of 5th Indian Conference on Computer Vision, Graphics and Image Processing. [ paper | talk ]
K.Zimmermann, T. Svoboda, J. Matas Multiview 3D Tracking with an Incrementally Constructed 3D Model, Third International Symposium on 3D Data Processing, Visualization and Transmission (3DPVT), Chapel Hill, USA, 2006. DEMO
Tomas Svoboda, Daniel Martinec, and Tomas Pajdla. A convenient multi-camera self-calibration for virtual environments. PRESENCE: Teleoperators and Virtual Environments, pp 407-422, 14(4), August 2005. MIT Press Manuscript of the journal article [ PDF, 1.7MB ]. Its bib-entry.
Petr Doubek, Indra Geys, Tomas Svoboda, Luc Van Gool: Cinematographic Rules Applied to a Camera Network. In Omnivis 2004. [PDF].
Hao Shao, Tomas Svoboda, Tinne Tuytelaars, and Luc Van Gool. Hpat indexing for fast object/scene recognition based on local appearance. In International Conference on Image and Video Retrieval, [PDF]. July 2003.
M. Gross, S. Wuermlin, M. Naef, E. Lamboray, Ch. Spagno, Kunz. A., E. Koller-Meier, T. Svoboda, L. Van Gool, S. Lang, Strehlke K., A. Vande Moere, and O. Staadt. Blue-c: A spatially immersive display and 3D video portal for telepresence. [PDF]. In Proceedings of ACM SIGGRAPH 2003, July 2003.
T. Svoboda, T. Pajdla, Epipolar Geometry for Central Catadioptric Cameras. International Journal of Computer Vision, 49(1):23-37, August 2002. [PDF]
Tomas Svoboda, Hanspeter Hug, and Luc Van Gool. ViRoom -- low cost synchronized multicamera system and its self-calibration. [PDF]. In Pattern Recognition, 24th DAGM Symposium, number 2449 in LNCS, pages 515-522. Springer, September 2002.
T.Svoboda, T.Pajdla. Matching in Catadioptric Images with Appropriate Windows and Outliers Removal. Proc. of the 9th International Conference on Computer Analysis of Images and Patterns. Number 2124 in Lecture Notes in Computer Science.
Tomas Svoboda. PhD Thesis: Central Panoramic Cameras, Design, Geometry, Egomotion. Center for Machine Perception, Czech Technical University.
Tomas Svoboda, Tomas Pajdla and Vaclav Hlavac. Epipolar Geometry for Panoramic Cameras. In 5th European Conference on Computer Vision, Freiburg, Germany, June 1998.
T. Svoboda and P. Sturm Badly Calibrated Camera in Ego-Motion Estimation - Propagation of Uncertainty, in 7-th International Conference Computer Analysis of Images and Patterns. 1997, Kiel, Germany.

PhD students

Patrik Vacek (PhD 2024, with KZ), [PhD thesis]
Zdenek Straka (PhD 2024, with M. Hoffmann) [PhD thesis]
Vladimir Kubelka (PhD 2018, with M. Reinstein) [PhD thesis awarded by Dean's prize for prestigious thesis]
Martin Pecka (PhD 2021, with KZ) [PhD Thesis]
Tomas Petricek (PhD 2018, with KZ) [PhD thesis]
David Hurych (PhD 2014, with K. Zimmermann) [PhD Thesis]
Karel Zimmermann (PhD 2008, Co-advised with J. Matas) [PhD thesis. Antonin Svoboda prize for the best PhD dissertation in the fields of cybernetics and informatics in 2008.]
Petr Doubek. (PhD at ETH, 2004; informally co-advised, advisor L. Van Gool.).

Professional Services

Reviewer for journals e.g. PAMI, IJCV, IJIG, JMIV and for major conferences like ICCV, CVPR, ECCV, and ACCV (ACCV2007 reviewer award). Area chair: AVSS2013,

Research fellowships

10/2000-10/2003, Switzerland, ETH (Swiss Federal Institute of Technology), Computer Vision Group headed by Prof. Luc Van Gool.
10-12/1998, Austria, Vienna University of Technology, PRIP group. Prof. Walter Kropatsch
3-7/1996, France, Grenoble, INRIA Rhone-Alpes. Research group MOVI. Prof. Roger Mohr
8-9/1994, Netherlands, Delft Technical University. Project Recognition of facial expresions. Prof. Leon Rothkrantz

Personal

CV [PDF]
students, collaborators. (not all, just from CMP)
Alumni Former students and collaborators. (not all, just from CMP)
Supporting