Mathematical model for obtaining a stereoscopic image from several wide-angle cameras of an aircraft
DOI:
https://doi.org/10.34185/1562-9945-1-138-2022-03Keywords:
FOV, stereo panorama, fisheye lens, distortion, stereo image, 360 ° video, VR, calibrationAbstract
A mathematical model for obtaining a stereo image from several wide-angle cameras was proposed. Methods for eliminating distortion from wide-angle “fisheye” lenses when constructing a stereo panorama were considered. It was shown that this approach was the most effective for performing key tasks.
References
C. Forster. Fast semi-direct monocular visual odometry [Text] / C. Forster, M. Pizzoli, and D. Scaramuzza // In IEEE Int. Conf. on Robotics and Automation. – 2014. – P. 15–22.
D. Scaramuzza. A flexible technique for accurate omnidirectional camera cali-bration and structure from motion [Text] / D. Scaramuzza, A. Martinelli, and R. Siegwart / Proceedings of the Fourth IEEE International Conference on Computer Vision Systems. – 2006, P. 45–45
D. Scharstein. Structure from motion using full spherical panoramic cameras [text] / D. Scharstein. A. Pagani and D. Stricker // In Int. Conf. on Computer Vision Workshops. –2011. P. 375–382.
W. Miled. Disparity map estimation using a total variation bound [Text] / W. Miled, J. C. Pesquet and M. Parent // Canadian Conf. Computer Robot Vision. –2006. P. 48–55.
D. Gallup. Real-Time Plane-Sweeping Stereo with Multiple Sweeping Directions [Text] / D. Gallup, J. Frahm, P. Mordohai, O. Yan, M. Pollefeys // Proceedings IEEE Conference on Computer Vision and Pattern Recognition. – 2007. – P. 1–8.
S. Im. All around depth from small motion with a spherical panoramic camera [Text] / S. Im, H. Ha, F. Rameau, H. Jeon, G. Choe, I. Kweon // Proceedings IEEE Conference on Computer Vision. – 2016. – P. 156–172.
C. Geyer. Catadioptric Projective Geometry [Text] / C. Geyer, K. Daniilidis. International Journal of Computer Vision. – 2001. – P. 223–243.
M. Schonbein. Omnidirectional 3d recon-struction in augmented manhattan worlds [Text] / M. Schonbein, A. Geiger. Proceedings IEEE Intelligent Robots and Systems. – 2014. – P. 716–723.
W. Gao. Dual-fisheye omnidirectional stereo [Text] / W. Gao, S. Shen. Proceedings IEEE Intelligent Robots and Systems. – 2017. – P. 6715–6722.
D. Scharstein. A taxonomy and evaluation of dense two-frame stereo corre-spondence algorithms [Text] / D. Scharstein, R. Szeliski, R. Zabih // Proceedings IEEE Workshop on Stereo and Multi-Baseline Vision. – 2001. – P. 131-140.
H. Hirschmuller. Evaluation of Stereo Matching Costs on Images with Radio-metric Differences [Text] / H. Hirschmuller, D. Scharstein // Proceedings IEEE Transactions on Pattern Analysis and Machine Intelligence. – 2009. – P. 1582-1599.
C. Forster. Semidirect Visual Odometry for Monocular and Multicamera Systems [Text] / C. Forster, Z. Zhang, M. Zhang, M. Werlberger, D. Scaramuzza // Proceedings IEEE Transactions on Robotics. – 2016. – P. 249–265.
Sweep line algorithm. [Electronic resource] //
URL: https://en.wikipedia.org/wiki/Sweep_line_algorithm.
T. Svoboda. Epipolar geometry for panoramic cameras [Text] / T. Svoboda, T. Pajdla, and V. Hlavac // In IEEE Conference on Computer Vision and Pattern Recog-nition. – 1998. P. 218–231.
H. Hirschmuller. Stereo processing by semiglobal matching and mutual infor-mation [Text] / H. Hirschmuller // Proceedings IEEE Transactions on pattern analy-sis and machine intelligence. – 2008. – P. 328–341.
