Project Details

Visual localization in natural environments (LOCATE)

Project Period: 1. 3. 2014 - 19. 7. 2016

Project Type: grant

Code: 4SGA8694

Agency: South Moravian Centre for International Mobility

Program:

Czech title
Vizuální lokalizace v přírodě
Type
grant
Keywords

image processing, computer graphics, computer vision, visual localization, terrain alignment, geo-localization, scene to model registration, viewing direction estimation, cameras, terrain mapping, cartographic data, photographic data, semantic image labeling, image registration

Abstract

The proposed project deals with localization in natural environments. It builds on the applicant's experience in registration of photos with a 3D geo-referenced terrain model and on the expertise of graphics group @FIT-BUT in image processing, computer graphics & vision, machine learning and experience with research and development of avionics. It proposes an investigation of image to model registration techniques allowing for building accurate visual localization systems that are more robust to qualities of input photos. Particular attention will be devoted to finding applications of visual localization systems in challenging scenarios, e.g. in aircraft systems.

Team members
Čadík Martin, doc. Ing., Ph.D. (UPGM FIT VUT) , research leader
Zemčík Pavel, prof. Dr. Ing. (UPGM FIT VUT) , team leader
Publications

2017

  • AHMAD Touqeer, CAMPR Pavel, ČADÍK Martin and BEBIS George. Comparison of Semantic Segmentation Approaches for Horizon/Sky Line Detection. In: Proceedings of the International Joint Conference on Neural Networks (IJCNN). Anchorage: Institute of Electrical and Electronics Engineers, 2017, pp. 4436-4443. ISBN 978-1-4799-1961-1. Detail
  • BREJCHA Jan and ČADÍK Martin. GeoPose3K: Mountain Landscape Dataset for Camera Pose Estimation in Outdoor Environments. Image and Vision Computing, vol. 2017, no. 1, pp. 1-41. ISSN 0262-8856. Detail
  • BREJCHA Jan and ČADÍK Martin. State-of-the-art in Visual Geo-localization. Pattern Analysis and Applications, vol. 2017, no. 3, pp. 1-25. ISSN 1433-7541. Detail

2016

  • PŘIBYL Bronislav, CHALMERS Alan, ZEMČÍK Pavel, HOOBERMAN Lucy and ČADÍK Martin. Evaluation of Feature Point Detection in High Dynamic Range Imagery. Journal of Visual Communication and Image Representation, vol. 38, no. 1, 2016, pp. 141-160. ISSN 1047-3203. Detail
  • ČADÍK Martin and AYDIN Tunc O. HDR Video Metrics. High Dynamic Range Video: Concepts, Technologies and Applications. London: Elsevier Science, 2016, pp. 111-125. ISBN 978-0-12-809477-8. Detail

2015

  • ČADÍK Martin, VAŠÍČEK Jan, HRADIŠ Michal, RADENOVIĆ Filip and CHUM Ondřej. Camera Elevation Estimation from a Single Mountain Landscape Photograph. In: British Machine Vision Conference 2015. Swansea: The British Machine Vision Association and Society for Pattern Recognition, 2015, pp. 1-12. ISBN 978-1-901725-53-7. Detail

2014

  • FIŠER Jakub, LUKÁČ Michal, JAMRIŠKA Ondřej, ČADÍK Martin, GINGOLD Yotam, ASENTE George and SÝKORA Daniel. Color Me Noisy: Example-based Rendering of Hand-colored Animations with Temporal Noise Control. Computer Graphics Forum, vol. 2014, no. 4, pp. 1-10. ISSN 0167-7055. Detail
Products

2014

  • Camera Pose Estimation from Line Correspondences using Plücker Coordinates, software, 2014
    Authors: Přibyl Bronislav, Zemčík Pavel, Čadík Martin Detail
Back to top