Tomas Pajdla

 

Associate Professor & Distinguished Researcher, Leader of AAG Group

AAG - Applied Algebra & Geometry Group

Czech Institute of Informatics, Robotics and Cybernetics

Czech Technical University in Prague

Visiting Associate Professor

National Institute of Informatics, Tokyo

 

 

 pajdla@cvut.cz  pajdla       

Get to us: Office B-638 CIIRC Building B Jugoslavskych partyzanu 3, 16000 Prague, Czechia

 

 

ELLIS Fellow

CVF Member

IEEE Member
SIAM Member

ECVA Board Member

ERC PE6 AdG Panel Member

NEURON Panel Member

 

 

Highlights

Research

ResTopics

Software

Projects

 

 

 

 

 

ICCV   2023  Best Paper Finalist

3DV    2022  General Chair

CVPR 2022  Best Paper

ICCV  2021  Best Paper Hnrbl Mention

ICCV  2019  Best Student Paper

 

see the past highlights

Interests  

Computer Vision, Applied Algebraic Geometry, Machine Learning, Robotics, Optimization

Papers  

See my ScholarDBLP & arXiv

Talks  

Solving Minimal Problems in the Age of Machine Learning @ GRASP Lab UPENN 2022, Minimal Problems in Computer Vision @ ICERM 2018,  Algebraic Geometry & 3D Reconstruction @ ISPRS 2016,
Solving Minimal Problems in Computer Vision Tutorial @ ICCV 2015,  Omnidirectional Vision Course @ ICCV 2003,
Stereo Geometries of Non-central Cameras 2001, How to write a good paper 2017    

Teaching  

Geometry of Computer Vision and Computer Graphics (CTU in Prague), Advanced Robot Kinematics (CTU in Prague),
Geometry of Computer Vision (Charles University in Prague), Geometric Problems in Robotics (Charles University in Prague), Geometry of Computer Vision (Ukrainian Catholic Uni in Lviv)

Links  

ELLIS, CVF, CMP, CIIRC Assembly, INRIA Willow, MAGIK Eye

 

Alumni

M Polic MSc 2015, PhD 2023 (CTU Prague)

P Hruby Bc 2019. MSc 2022 (ETH Zurich)

C Albl MSc 2011, PhD 2019 (ETH Zurich)

V Korotynskiy Bc, MSc 2020 (CTU Prague)
S Steidl MSc 2018 (CTU Prague)

P Trutman MSc/Bc 2018 (CTU Prague)
P Gronat
 PhD 2017 (AVAST)

O Rybkin Bc 2017 (U Penn)

J Heller PhD 2016 (MAGIK Eye)

F Srajer MSc 2016 (ETH Zurich)

V Smutny PhD 2016 (CTU Prague)

M Jancosek PhD 2015 (Capturing Reality)

Z Kukelova PhD 2013 (Microsoft Resrch)

M Bujnak PhD 2013 (Capturing Reality)

M Havlena PhD 2012 (ETH Zurich)

J Smisek MSc 2011 (Delft University)

J Knopp MSc 2009 (KU Leuven)

T Ehlgen PhD 2008 (Bosch)

H Bakstein PhD 2006 (EU Patent Office)

B Micusik PhD 2004 (Stanford University)

D Martinec PhD 2003 (Microsoft)

J Sivic MSc 2002 (Oxford University)

O Chum MSc 2001 (CTU in Prague)

R Horcik MSc 2001 (CTU in Prague)

T Svoboda PhD 2000 (ETH Zurich)

 

 

Research

Using monodromy to recover symmetries of polynomial systems

T Duff, V Korotynskiy, T Pajdla, M Regan

ISSAC 2023 | pdf

An efficient model for a camera behind a parallel refractive slab

A Lasaruk, T Pajdla

International Journal of Computer Vision 2023 | pdf

Viewing Graph Solvability in Practice

F Arrigoni, T Pajdla, A Fusiello

ICCV 2023 The Best Paper Finalist | pdf

Four-view geometry with unknown radial distortion

P Hruby, V Korotynskiy, T Duff, L Oeding, M Pollefeys, T Pajdla, V Larsson

CVPR 2023 | pdf | supp|github

Galois/monodromy groups for decomposing minimal problems in 3D reconstruction

T Duff, V Korotynskiy, T Pajdla, M H Regan

SIAM Journal on Applied Algebra and Geometry 2022 | arXiv

 

Trifocal relative pose from lines at points

R Fabbri, T Duff, H Fan, M Regan, D da Costa de Pinho, E Tsigaridas, C Wampler, J Hauenstein, P J Giblin, B B Kimia, A Leykin, T Pajdla

IEEE TPAMI 2022 | pdf

 

Objects can move: 3D change detection by geometric transformation consistency

A Adam, T Sattler, K Karantzalos, T Pajdla

ECCV 2022 | arXiv

Revisiting viewing graph solvability: an effective approach based on cycle consistency

F Arrigoni, A Fusiello, R Rizzi, E Ricci, T Pajdla

IEEE TPAMI 2022 | pdf

D-InLoc++: Indoor Localization in Dynamic Environments

M Dubenova, A Zderadickova, O Kafka, T Pajdla, M Polic

DAGM German Conference on Pattern Recognition 2022 | arXiv

Improving image pair selection for large scale Structure from Motion by introducing modified Simpson coefficient

T Kato, I Shimizu, T Pajdla

IEICE TRANSACTIONS on Information and Systems 2022 | pdf

 

 

Globally optimal solution to inverse kinematics of 7DOF serial manipulator

P Trutman, M Safey El Din, D Henrion, T Pajdla

IEEE Robotics and Automation Letters 2022 | pdf | arXiv | ICRA 2023 (oral)

 

Multi-frame motion segmentation by combining two-frame results

F Arrigoni, E Ricci, T Pajdla

International Journal of Computer Vision 2022 | pdf

 

Automatic self-contained calibration of an industrial dual-arm robot with cameras using self-contact, planar constraints, and self-observation

K Stepanova, J Rozlivek, F Puciow, P Krsek, T Pajdla, M Hoffmann

Robotics and Computer-Integrated Manufacturing 2022 | pdf

 

Optimizing elimination templates by greedy parameter search

E Martyushev, J Vrablikova, T Pajdla

CVPR 2022 | pdf | supp | github

 

Learning to solve hard minimal problems

P Hruby, T Duff, A Leykin, T Pajdla

CVPR 2022 (oral) The Best Paper | pdf | supp | arXiv | github

 

Motion segmentation with pairwise matches and unknown number of motions

F Arrigoni, L Magri, T Pajdla

ICPR 2021 | pdf

 

Viewing graph solvability via cycle consistency

F Arrigoni, A Fusiello, E Ricci, T Pajdla

ICCV 2021 (oral) The Best Paper Honorable Mention | pdf | supp

 

A picture containing text

Description automatically generated

Long-Term Visual Localization Revisited
C Toft, W Maddern, A Torii, L Hammarstrand, E Stenborg, D Safari, M Okutomi, M Pollefeys,

J Sivic, T Pajdla, F Kahl, T Sattler

IEEE TPAMI 2020 | pdf

 

A picture containing different

Description automatically generated

Minimal Rolling Shutter Absolute Pose with Unknown Focal Length and Radial Distortion

Z Kukelova, C. Albl, A Sugimoto, T Pajdla, K Schindler

ECCV 2020 (spotlight) | pdf | supp

 

A picture containing text, music

Description automatically generated

PL1P – Point-line Minimal Problems under Partial Visibility in Three Views
T Duff, K Kohn, A Leykin, T Pajdla

ECCV 2020 | pdf | supp | github

 

On the Usage of the Trifocal Tensor in Motion Segmentation

F Arrigoni, L Magri, T Pajdla

ECCV 2020 | pdf | supp

Diagram

Description automatically generated

Making Affine Correspondences Work in Camera Geometry Computation

D Barath, M Polic,  Z Kukelova, T Sattler, W Forstner, T Pajdla

ECCV 2020 | pdf | supp

 

NCNet: Neighbourhood Consensus Networks for Estimating Image Correspondences
I Rocco, M Cimpoi, R Arandjelovic, A Torii, T Pajdla, J Sivic
IEEE TPAMI 2020 pdf | NeurIPS 2018 pdf arXiv | Project page

A picture containing text, sky, colorful

Description automatically generated

Rolling Shutter Camera Absolute Pose

C Albl, Z Kukelova, V Larsson, T Pajdla

IEEE PAMI 42:(6) 1439-1452, 2020 | pdf | github

 

A screenshot of a computer

Description automatically generated with medium confidence

From Two Rolling Shutters to One Global Shutter

C Albl, Z Kukelova, V Larsson, M Polic, T Pajdla, K Schindler

CVPR 2020 (oral) | pdf | supp | arXiv | github

A picture containing indoor, computer, different, various

Description automatically generated

TRPLP – Trifocal Relative Pose from Lines at Points
R Fabbri, T Duff, H Fan, M H Regan, D da Costa de Pinho, E Tsigaridas, C W Wampler, J D Hauenstein, P J Giblin, B Kimia, A Leykin, T Pajdla

CVPR 2020 | paper | supp | arXiv | Minus @ github

A collage of a robot

Description automatically generated with medium confidence

Robot Self-Calibration Using Multiple Kinematic Chains - A Simulation Study on the iCub Humanoid Robot
K Stepanova, T Pajdla, M Hoffmann
IEEE Robotics Automation Letters 2019 |
IEEE Explore | pdf

PLMP – Point-Line Minimal Problems in Complete Multi-View Visibility

T Duff, K Kohn, A Leykin, T Pajdla

ICCV 2019 | paper | arXiv | github (oral) The Best Student Paper  | TPAMI 2023

A picture containing grass, photo, many

Description automatically generated

Robust Motion Segmentation from Pairwise Matches.

F Arrigoni, T Pajdla

ICCV 2019 | paper | arXiv

A picture containing indoor, wall

Description automatically generated

Is This the Right Place? Geometric-Semantic Pose Verification for Indoor Visual Localization

H Taira, I Rocco, J Sedlar, M Okutomi, J Sivic, T Pajdla, T Sattler, A Torii

ICCV 2019 | paper | arXiv | Project page

D2-Net: A Trainable CNN for Joint Description and Detection of Local Features

M Dusmanu, I Rocco, T Pajdla, M Pollefeys, J Sivic, A Torii, T Sattler

CVPR 2019 | pdf | Supplementary material | Project page

Linear solution to the minimal absolute pose rolling shutter problem

Z Kukelova, C Albl, A Sugimoto, T Pajdla
ACCV 2018 | paper |arXiv

Fast and Accurate Camera Covariance Computation for Large 3D Reconstruction
M Polic, W Förstner, T Pajdla
ECCV 2018 | paper |arXiv

Beyond Gröbner Bases: Basis Selection for Minimal Solvers

V Larsson, M Oskarsson, K Åström, A Wallis, Z Kukelova, T Pajdla
CVPR 2018
pdf | paper |arXiv

InLoc: Indoor Visual Localization with Dense Matching and View Synthesis

H Taira, M Okutomi, T Sattler, M Cimpoi, M Pollefeys, J Sivic, T Pajdla, A Torii

CVPR 2018 pdf | arXiv | Project page | bib

Benchmarking 6DOF Urban Visual Localization in Changing Conditions

T Sattler, W Maddern, C Toft, A Torii, L Hammarstrand, E Stenborg, D Safari, M Okutomi, M Pollefeys, J Sivic, F Kahl, T Pajdla
CVPR 2018
pdf | arXiv | Project page

24/7 Place Recognition by View Synthesis
A Torii, R Arandjelovic, J Sivic, M Okutomi, T Pajdla
IEEE TPAMI 2018 pdf | CVPR 2015 pdf | Project page+data+code

 

NetVLAD: CNN architecture for weakly supervised place recognition

R Arandjelovic, P Gronat, A Torii, T Pajdla and J Sivic

IEEE PAMI 2018 pdf | CVPR 2016 pdf | Project page (oral) | video

 

Camera Uncertainty Computation in Large 3D Reconstruction

M Polic, T Pajdla
3DV 2017 pdf | Project page

Nautilus: recovering regional symmetry transformations for image editing

M Lukac, D Sykora, K Sunkavalli, E Shechtman, O Jamriska, N Carr, T Pajdla

SIGGRAPH 2017 | ACM TOG Journal PDF | Project page

 

Are Large-Scale 3D Models Really Necessary for Accurate Visual Localization?

T Sattler, A Torii, J Sivic, M Pollefeys, H Taira, M Okutomi, T Pajdla

CVPR 2017 | pdf | Project page | TPAMI 2019

 

On the Two-View Geometry of Unsynchronized Cameras

C Albl, Z Kukelova, A Fitzgibbon, J Heller, M Smid, T Pajdla

CVPR 2017 | pdf | Supplementary | github

 

A clever elimination strategy for efficient minimal solvers

Z Kukelova, J Kileel, B Sturmfels, T Pajdla

CVPR 2017 | pdf | Supplementary | arXiv

 

Distortion varieties

J Kileel, Z Kukelova, T Pajdla, B Sturmfels

Foundations of Computational Mathematics | arXiv

 

Degeneracies in rolling shutter SfM

C Albl, A Sugimoto, T Pajdla

ECCV 2016 | LNCS | pdf

 

Globally optimal hand-eye calibration using branch-and-bound

J Heller, M Havlena, T Pajdla

IEEE TPAMI 2016 | pdf

Learning and Calibrating Per-Location Classifiers for Visual Place Recognition

P Gronat, J Sivic, G Obozinski, T Pajdla

IJCV 2016 pdf | CVPR 2013 pdf

 

GpoSolver: a Matlab/C++ toolbox for global polynomial optimization

J Heller, T Pajdla

Optimization Methods and Software 2016 | pdf | Software

 

 

Software from AAG

Eficient Minimal Solver Templates

Radial Quadrifocal Tensor

Learning to start Homotopy Continuation

Two-Body SfM

Minimal problem classification in M2

Automatic Generator of Minimal Solvers

YASM – Yet Another SfM

CMP SfM Webservice (AliceVision)
CMPMVS  – MV Stereo (
AliceVision)

GpoSolver – Global Polynom Optimization

radhomo-0.1 – Rad Dist Homogrphy [pdf]

F10-0.1 – Rad Relative Pose [pdf]

bbhec – BB Hand-Eye Optimization

mpherwc – Hand-Eye & R-W Cal [pdf]

minhec – Hand-Eye Cal Minimal [pdf]

polyopt  –  Polynom Opt by SDP Relax

TE-inversion – Uncertainty propag in SfM

 

A close - up of a statue

Description automatically generated with low confidence

 

Projects

2019 SPRING project (H 2020 EU)

2019 ARtwin project (H 2020 EU)

2017  

2017 IMPACT Project (EU & MSMT)

2017  & CTU Research Lab

2016 LADIO Project (H2020 EU)

 

 

 

Student Research Topics

Research topics for PhD, MSc and BSc students.

 

3D Scene Reconstruction from Images

3D scene reconstruction from images is a fundamental problem of computer vision. It finds many applications in industry ranging from autonomous driving to movie special effects. The topic is best for students with interest in algorithms, experimental work, and engineering of really working systems.
Algebraic Methods in Computer Vision and Robotics

Algebraic techniques have proved very useful in solving difficult problems in geometry of computer vision. We will aim at studying more advanced elements of algebraic geometry and applying them to real engineering problems. The topic is best for students with interest in applied mathematics.
Image-based Scene Recognition and Visual Localization

Visual scene recognition and image-based localization is an important problem in computer vision and machine learning. We will aim developing new approaches to place representation and its search. The topic is suitable for students with interest computer vision and machine learning applied to real engineering problems.

Polynomial Optimization in Computer Vision and Robotics

Polynomial optimization techniques proved very useful in solving interesting problems in geometry of computer vision and robotics. We will aim at studying more polynomial optimization techniques and applying them in computer vision and robotics. The topic is suitable for students with interest in applied mathematics but used on real engineering problems.

 

 

Past News

2020 Nov ACCV 2020 Program Chair, 2020 Sep ELLIS Unit @ CIIRC established, 2020 Sep ECCV 2020 4 papers (1 spotlight), 2020 Jun CVPR 2020 3 papers (1 oral), 2019 Oct  ICCV 2019 1 oral + 2 papers, 2019 Oct  ICCV 2019 3DRW workshop, 2019 Aug SPRING H2020 project granted, 2019 Jun  CVPR 2019 paper, 2019 Mar ARtwin H2020 Project granted, 2018 December ELLIS Founded, Papers @ NeurIPS 2018 & ACCV 2018, 2018 Sep-Dec. 2019 Feb/May, Visiting ICERM @ Brown University, 2018 September Paper @ ECCV 2018, 2018 June 3 Papers @ CVPR 2018, 2018 June  Minimal Problems Tutorial, 2018 June CVPR 2019 Workshop on Autonomous driving, 2018 June CVPR 2019 Workshop on LT Visual Localization ,2018 March GC of 3DRW Workshop @ ECCV 2018, 2017 September Continental project, 2017 September PC of 3DV 2018,  2017 July J Heller received CSKI Prize 4 his PhD, 2017 July 3 Papers @ CVPR 2017, 2017 June IMPACT project with INRIA started,  2017 May MAGIK Eye – CTU Common Lab founded, 2016 November PC of ACCV 2020, June 2016 3 papers @ CVPR 2016, June 2015 3 papers @ CVPR 2015, January 2015 LADIO EU Project starts.

 

©Tomas Pajdla 2017-2024