Czech Technical University Prague

Czech Technical University Prague

The Center for Machine Perception @ the Czech Technical University Prague

Prague, CZ


The Centre for Machine Perception (CMP) at the Department of Cybernetics of the Czech Technical University in Prague is a computer vision, pattern recognition, and mathematics of uncertainty research unit. The expertise of the CMP includes fundamentals as well as applications of computer vision, pattern recognition and machine learning including calibrating multi-camera optical systems, stereo reconstruction, omnidirectional and panoramic cameras based on mirrors and lenses, image matching and object recognition in images. Since 1995, CMP closely collaborates with its spin-off company Neovision s.r.o. on developing computer vision applications for industry.

In PRoViScout, CMP will concentrate on applying computer vision technology to planetary robotics including developing apropriate camera models, calibration procedures, visualizations and image matching for 3D reconstruction. CTU is reponsible for mission status visualisation over the Internet and for managing disemination activities.

For more information, visit the webpage of T. Pajdla.

Team Members


Dr. Tomas Pajdla

Tomas Pajdla is an assistant professor at the Czech Technical University Prague. He works in geometry and algebra of computer vision, eye-hand calibration, precise digital optical measurements, photogrammetry, robot navigation using vision, image matching and object recognition. He co-authored works that introduced epipolar geometry of panoramic and non-central cameras, investigated the use of omnidirectional images in structure from motion and camera pose computation, contributed to studies of panoramic mosaics, contributed to solving the image matching and stereo correspondence problem, and to algebraic techniques in computer vision applied to solving minimal problems. He published more than 50 works in scientific journals and refereed conferences and was awarded prizes at OAGM 1998, BMVC 2002 and ICCV camera localization contest in 2005. T. Pajdla served as a Program Chair of the 8th European Conference on Computer Vision 2004 in Prague and is a member of the ECCV Board. He acted as an area chair of ICCV 2005, 2009, CVPR 2006, 2007 ACCV 2006, 2007, and BMVC 2005-2008 and was a coeditor of a special issue of the International Journal on Computer Vision. T. Pajdla received the Outstanding Reviewer Award at the CVPR 2009.

For more information, visit the webpage of
T. Pajdla.


RNDr. Michal Jancosek

Michal Jancosek is a PhD student at the CMP. His expertise is in image matching and multiview stereo.

For more information, visit the webpage of
M. Jancosek.


Ing. Petr Gronat

Petr Gronat is a PhD student at the CMP. He is working on image-based localization and place recognition.

For more information, visit the webpage of
P. Gronat.