Homography Estimation Between Omnidirectional Cameras Without Point Correspondences

  • Robert Frohlich
  • Levente Tamás
  • Zoltan KatoEmail author
Part of the Studies in Systems, Decision and Control book series (SSDC, volume 42)


This chapter presents a novel approach for homography estimation between omnidirectional cameras. The solution is formulated in terms of a system of nonlinear equations. Each equation is generated by integrating a nonlinear function over corresponding image regions on the surface of the unit spheres representing the cameras. The method works without point correspondences or complex similarity metrics, using only a pair of corresponding planar regions extracted from the omnidirectional images. Relative pose of the cameras can be factorized from the estimated homography. The efficiency and robustness of the proposed method has been confirmed on both synthetic and real data.


Synthetic Dataset Surface Patch Segmentation Error Spherical Image Camera Center 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



This research was partially supported by Domus MTA Hungary; and by the European Union and the State of Hungary, co-financed by the European Social Fund through projects (grant no.: TAMOP-4.2.2.C-11/1/KONV-2012-0013) and TAMOP-4.2.4.A/2-11-1-2012-0001 National Excellence Program. The authors would like to thank Levente Hajder for the Matlab implementation of the factorization method from Faugeras and Lustman (1988).


  1. Baker S, Nayar SK (1999) A theory of single-viewpoint catadioptric image formation. Int J Comput Vis 35(2):175–196CrossRefGoogle Scholar
  2. Basri R, Jacobs DW (1996) Recognition using region correspondences. Int J Comput Vis 25:141–162Google Scholar
  3. Caron G, Marchand E, Mouaddib EM (2011) Tracking planes in omnidirectional stereovision. In: IEEE international conference on robotics and automation. IEEE, pp 6306–6311Google Scholar
  4. Coughlan J, Yuille AL (1999) Manhattan world: compass direction from a single image by bayesian inference. In: The Proceedings of the Seventh IEEE International Conference on Computer Vision, 1999, vol 2, pp 941–947. doi: 10.1109/ICCV.1999.790349
  5. Domokos C, Nemeth J, Kato Z (2012) Nonlinear shape registration without correspondences. IEEE Trans Pattern Anal Mach Intell 34(5):943–958. doi: 10.1109/TPAMI.2011.200 CrossRefGoogle Scholar
  6. Faugeras O, Lustman F (1988) Motion and structure from motion in a piecewise planar environment. Technical Report RR-0856, INRIA, Sophia Antipolis, France.
  7. Furukawa Y, Curless B, Seitz S, Szeliski R (2009) Manhattan-world stereo. In: IEEE conference on computer vision and pattern recognition. IEEE Computer Society, Los Alamitos, pp 1422–1429. doi: 10.1109/CVPRW.2009.5206867
  8. Geyer C, Daniilidis K (2000) A unifying theory for central panoramic systems. In: European conference on computer vision (ECCV), pp 445–462Google Scholar
  9. Gutierrez D, Rituerto A, Montiel J, Guerrero J (2011) Adapting a real-time monocular visual slam from conventional to omnidirectional cameras. In: 2011 IEEE international conference on computer vision workshops (ICCV Workshops), pp 343–350. doi: 10.1109/ICCVW.2011.6130262
  10. Hartley R, Zisserman A (2003) Multiple view geometry in computer vision, 2nd edn. Cambridge University Press, New YorkGoogle Scholar
  11. Kannala J, Brandt SS (2006) A generic camera model and calibration method for conventional, wide-angle, and fish-eye lenses. IEEE Trans Pattern Anal Mach Intell 28(8):1335–1340CrossRefGoogle Scholar
  12. Lowe DG (2004) Distinctive image features from scale-invariant keypoints. Int J Comput Vis 60(2):91–110CrossRefGoogle Scholar
  13. Makadia A, Geyer C, Daniilidis K (2007) Correspondence-free structure from motion. Int J Comput Vis 75(3):311–327. doi: 10.1007/s11263-007-0035-2 CrossRefGoogle Scholar
  14. Mei C, Rives P (2007) Single view point omnidirectional camera calibration from planar grids. IEEE international conference on robotics and automation (ICRA). Roma, Italy, pp 3945–3950Google Scholar
  15. Mei C, Benhimane S, Malis E, Rives P (2008) Efficient homography-based tracking and 3-D reconstruction for single-viewpoint sensors. IEEE Trans Robot 24(6):1352–1364. doi: 10.1109/TRO.2008.2007941 CrossRefGoogle Scholar
  16. Mičušík B, Pajdla T (2004) Para-catadioptric camera auto-calibration from epipolar geometry. In: Hong KS, Zhang Z (eds) Proceedings of the asian conference on computer vision (ACCV). Asian Federation of Computer Vision Societies, Seoul, Korea South, vol 2, pp 748–753Google Scholar
  17. Molnár J, Frohlich R, Dmitry C, Kato Z (2014a) 3D reconstruction of planar patches seen by omnidirectional cameras. Proceedings of international conference on digital image computing: techniques and applications. IEEE, Wollongong, Australia, pp 1–8Google Scholar
  18. Molnár J, Huang R, Kato Z (2014b) 3D reconstruction of planar surface patches: A direct solution. In: Jawahar CV, Shan S (eds) Proceedings of ACCV workshop on big data in 3D computer vision, Lecture notes in computer science, vol 9008. Springer, Singapore, pp 286–300Google Scholar
  19. Nayar SK (1997) Catadioptric omnidirectional camera. In: Proceedings of the 1997 conference on computer vision and pattern recognition (CVPR ’97). IEEE Computer Society, Washington, USA, CVPR ’97, pp 482–488.
  20. Puig L, Guerrero JJ (2011) Scale space for central catadioptric systems: Towards a generic camera feature extractor. In: Proceedings of international conference on computer vision. IEEE, pp 1599–1606Google Scholar
  21. Puig L, Guerrero JJ (2013) Omnidirectional vision systems: calibration, feature extraction and 3D information. SpringerGoogle Scholar
  22. Puig L, Bastanlar Y, Sturm P, Guerrero J, Barreto J (2011) Calibration of central catadioptric cameras using a DLT-Like approach. Int J Compu Vis 93(1):101–114. doi: 10.1007/s11263-010-0411-1,
  23. Saurer O, Fraundorfer F, Pollefeys M (2012) Homography based visual odometry with known vertical direction and weak Manhattan world assumption. In: IEEE/IROS workshop on visual control of mobile robots (ViCoMoR)Google Scholar
  24. Scaramuzza D, Martinelli A, Siegwart R (2006a) A flexible technique for accurate omnidirectional camera calibration and structure from motion. In: Proceedings of the fourth IEEE international conference on computer vision systems. IEEE Computer Society, Washington, USA, ICVS-06, pp 45–51Google Scholar
  25. Scaramuzza D, Martinelli A, Siegwart R (2006b) A toolbox for easily calibrating omnidirectional cameras. In: Proceedings of the IEEE/RSJ international conference on intelligent robots. IEEE, Beijing, pp 5695–5701Google Scholar
  26. Sturm P (2000) Algorithms for plane-based pose estimation. Proc Int Conf Comput Vis Pattern Recognit 1:706–711. doi: 10.1109/CVPR.2000.855889 Google Scholar
  27. Sturm P, Ramalingam S, Tardif JP, Gasparini S, Barreto J (2011) Camera models and fundamental concepts used in geometric computer vision. Found Trends Comput Graph Vis 6(1–2):1–183. doi: 10.1561/0600000023,

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Institute of InformaticsUniversity of SzegedSzegedHungary
  2. 2.Department of AutomationTechnical University of Cluj-NapocaCluj-NapocaRomania
  3. 3.Department of Mathematics and InformaticsJ. Selye UniversityKomarnoSlovakia

Personalised recommendations