Direct Fisheye Stereo Correspondence Using Enhanced Unified Camera Model and Semi-Global Matching Algorithm

Bogdan Khomutenko (1, 2) , Gaëtan Garcia (3) , Philippe Martinet (4)

1 ECN - École Centrale de Nantes
2 LS2N - Laboratoire des Sciences du Numérique de Nantes
3 IRCCyN - Institut de Recherche en Communications et en Cybernétique de Nantes
4 LASMEA - Laboratoire des sciences et matériaux pour l'électronique et d'automatique

In this paper, it is shown that the Enhanced Unified Camera Model, which is used to model fisheye cameras, projects straight lines into conic sections. Then a way to find equations of straight line projections is proposed. The method is applied to epipolar curves to adapt the Semi-Global Matching algorithm to fisheye stereo systems to compute a dense direct stereo correspondence without undistortion and rectification of fisheye images. It is done by efficient image sampling along an epipolar curve, using an algorithm for rasterization of implicit curves. A C++ implementation is available online.

Domaines: Robotique [cs.RO]; Automatique / Robotique; Vision par ordinateur et reconnaissance de formes [cs.CV]

Bogdan Khomutenko, Gaëtan Garcia, Philippe Martinet. Direct Fisheye Stereo Correspondence Using Enhanced Unified Camera Model and Semi-Global Matching Algorithm. ICARCV 2016, Nov 2016, Phuket, Thailand. ⟨hal-01722268⟩