crta
Hrvatska znanstvena Sekcija img
bibliografija
3 gif
 Naslovna
 O projektu
 FAQ
 Kontakt
4 gif
Pregledavanje radova
Jednostavno pretraživanje
Napredno pretraživanje
Skupni podaci
Upis novih radova
Upute
Ispravci prijavljenih radova
Ostale bibliografije
Slični projekti
 Bibliografske baze podataka

Pregled bibliografske jedinice broj: 954083

Zbornik radova

Autori: Jovan Mijatov; Vladimir Zlokolica; Ratko Obradović; Relja Ljubobratović; Radovan Obradović; Vladimir Mastilović; Krešimir Vdovjak; Irena Galić
Naslov: Multi-Camera System Calibration for 3D Scene Reconstruction, Modeling and Visualization
Izvornik: Conference Proceedings of 6th International Conference on Geometry and Graphics (MONGEOMETRIJA 2018)
Skup: The 6th International Conference on Geometry and Graphics (MONGEOMETRIJA 2018)
Mjesto i datum: Srbija, Novi Sad, 6.-9.6.2018.
Ključne riječi: camera calibration ; computer vision ; photogrammetry ; 3D modeling ; multi-view stereo ; structure-from-motion, architecture ; design technology ; fabrication ; education
Sažetak:
Current state-of-the-art multi-camera-view 3D reconstruction algorithms provide high quality 3D point cloud, which is further used for 3D meshing, texture mapping and finally generating corresponding 3D model of the scene. One of the most important factors for both 3D surface reconstruction and texture mapping is accurate and robust camera calibration. This paper presents a novel approach for multicamera system calibration where we extend and make mixture of known calibration techniques to calibrate multi-camera system with specially designed three-dimensional calibration object. More precisely, we extend calibration technique that uses planar pattern to three-dimensional pattern camera calibration. First, camera parameters are roughly estimated for all cameras in multiple camera system, then these parameters are refined using standard bundle adjustment algorithm. The proposed approach has been tested with synthetic and real data. Mean square error between re-projected and detected pattern points has been found to be in the range of 0.1px – 0.5px, for real camera data of 24MPx. The proposed method is shown to be more accurate or comparable to other calibration techniques.
Vrsta sudjelovanja: Predavanje
Vrsta prezentacije u zborniku: Cjeloviti rad (više od 1500 riječi)
Vrsta recenzije: Međunarodna recenzija
Izvorni jezik: ENG
Kategorija: Znanstveni
Znanstvena područja:
Računarstvo
Upisao u CROSBI: Krešimir Vdovjak (kresimir.vdovjak@etfos.hr), 14. Ruj. 2018. u 07:39 sati



Verzija za printanje   za tiskati


upomoc
foot_4