A study of Kruppa’s equation for camera self-calibration

Publikationstyp
Conference Paper
Date Issued
2014
Sprache
English
Author(s)
Prapitasari, Luh Putu Ayu  
Grigat, Rolf-Rainer  
Institut
Bildverarbeitungssysteme E-2  
TORE-URI
http://hdl.handle.net/11420/9932
Citation
Proceedings of the International Conference of Machine Vision and Machine Learning Paper No. 57: 1-8 (2014)
Contribution to Conference
International Conference of Machine Vision and Machine Learning, (MVML '14)  
In this paper we discuss the Kruppa’s equation, which is widely known as the first camera self-calibration method. The classical and the simplified version of the method are of our interest. We have analyzed it based on how it is derived. As Kruppa’s equation works based on the image of the absolute conic, which is an imaginary conic, we have found that the use of circular points at infinity was missed in the classical Kruppa’s in generating the imaginary conic. We have pointed out that the concept of degenerate point conic in deriving the simplified version is one of the reasons of degeneracy. Other than that, we also found some other inaccuracies of the derivation. For the future use as well as the improvement of the method, we have proposed possible solutions for it.
Subjects
Self-calibration
Kruppa’s equation
Absolute conic
Circular points at infinity
DDC Class
004: Informatik