3D SCENE RECONSTRUCTION BASED ON CAMERA MOTION
DOI:
https://doi.org/10.24867/19BE01KudusKeywords:
3D reconstruction, Structure-from-Motion, aerial photographyAbstract
This paper describes the process of obtaining 3D point cloud using the set of photographs taken from a drone. Relationship between 3D coordinates of the point and corresponding 2D projections as well as a description of Structure-from-Motion (SfM) algorithm is given. Also, an approach for overcoming the problem of measurements based on aerial photography, by using obtained 3D model, is discussed.
References
[1] W. Förstner, B. Wrobel, “Photogrammetric Computer Vision: Statistics, Geometry, Orientation and Reconstruction”, Springer, 2016
[2] R. Hartley , A. Zisserman,“Multiple View Geometry in Computer Vision“, Cambridge UP, 2004
[3] https://www.e-education.psu.edu/natureofgeoinfo/book/export/html/1815 (pristupljeno 01.2022)
[4] L. Abrahamczyk , J. Schwatz , “Forecast Engineering: From Past Design to Future Decision”, Bauhaus-Universität Weimar, Fakultät Bauingenieurwesen, Institut für Konstruktiven Ingenieurbau, 2017
[5] https://www.agisoft.com/ (pristupljeno 01.2022)
[6] https://verticalphotoplacer.github.io/VerticalPhotoPlacer/ (pristupljeno 01.2022)
[2] R. Hartley , A. Zisserman,“Multiple View Geometry in Computer Vision“, Cambridge UP, 2004
[3] https://www.e-education.psu.edu/natureofgeoinfo/book/export/html/1815 (pristupljeno 01.2022)
[4] L. Abrahamczyk , J. Schwatz , “Forecast Engineering: From Past Design to Future Decision”, Bauhaus-Universität Weimar, Fakultät Bauingenieurwesen, Institut für Konstruktiven Ingenieurbau, 2017
[5] https://www.agisoft.com/ (pristupljeno 01.2022)
[6] https://verticalphotoplacer.github.io/VerticalPhotoPlacer/ (pristupljeno 01.2022)
Downloads
Published
2022-09-05
Issue
Section
Electrotechnical and Computer Engineering
