Abstract
This article describes an original method to reconstruct a 3D scene from a sequence of images. Our approach uses both the dense 3D point cloud extracted by multi-view stereovision and the calibrated images. It combines depth-maps construction in the image planes with surface reconstruction through restricted Delaunay triangulation. The method may handle very large scale outdoor scenes. Its accuracy has been tested on numerous outdoor scenes including the dense multi-view benchmark proposed by Strecha et al. Our results show that the proposed method compares favorably with the current state of the art.
Highlights
Various applications in Computer Vision, Computer Graphics and Computational Geometry require a surface reconstruction from a 3D point cloud extracted by stereovision from a sequence of overlapping images
Tracks with small cone angles 2 are discarded. Note that this criterion is very useful for filtering 3D point clouds extracted from a set of images taken from a video sequence
To show the ability of our algorithm to cope with much bigger outdoor scenes and its ability to produce meshes that take into account the user budget for the size of the output mesh, we have chosen to test our method on the Aiguille du Midi dataset (©B.Valet/IMAGINE)
Summary
Various applications in Computer Vision, Computer Graphics and Computational Geometry require a surface reconstruction from a 3D point cloud extracted by stereovision from a sequence of overlapping images. Work on reconstruction has been focused on data acquired through laser devices, with characteristics that the obtained 3D point cloud is dense and well distributed [1]-[3]. These reconstruction methods have been reported as successful, the nature of the laser scanners greatly limits their usefulness for large-scale outdoor reconstructions. On top of these point clouds, the additional information provided by the calibrated images can be exploited to help surface reconstruction
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
