Dense and Accurate Multiview Stereo Reconstruction Using Differential Geometry of Curves/Surfaces, Lofting and Occlusion Reasoning

Usumezbas, Anil

Full Metadata

Overview

Year:: 2017
Contributor:: Usumezbas, Anil (creator); Kimia, Benjamin (Advisor); Fabbri, Ricardo (Reader); Taubin, Gabriel (Reader); Brown University. School of Engineering (sponsor)
Genre:: theses
Subject:: 3D vision; Geometry, Differential; Three-dimensional modeling; structure from motion; multiview stereo
Extent:: xix, 85 p.
DOI: https://doi.org/10.7301/Z06H4FWN

Files

Description

Abstract:: Reconstructing 3D scenes from multiple views has made impressive strides in recent years, chiefly by correlating isolated feature points, intensity patterns, or curvilinear structures. In the general setting – without controlled acquisition, abundant texture, curves and surfaces following specific models or limiting scene complexity – most methods produce unorganized point clouds, meshes, or voxel representations, with some exceptions producing unorganized clouds of 3D curve fragments or surface patches. Ideally, many applications require structured representations of curves, surfaces and their spatial relationships. The work presented in this thesis is a step in this direction by formulating an approach that combines 2D image curves into a collection of 3D curves, with topological connectivity between them represented as a 3D graph. This results in a 3D drawing, which is complementary to surface representations in the same sense as a 3D scaffold complements a tent taut over it. Furthermore, this 3D graph serves as a boundary condition for a novel dense surface reconstruction algorithm based on occlusion reasoning and lofting, a computer graphics method for curve interpolation. Because reconstructed surface representations are constrained by the 3D drawing acting like a scaffold to hang on the computed representations, the resulting surfaces are consistent with the curve boundaries of the 3D drawing, and do not smooth over important features such as ridges or corners. The 3D models obtained this way are capable of addressing some of the shortcomings of the multiview stereo state-of- the-art. We evaluate all our results against ground truth on synthetic and real datasets.
Notes:: Thesis (Ph. D.)--Brown University, 2017

Content

Access Conditions

Rights: In Copyright
Restrictions on Use: Collection is open for research.

Citation

Usumezbas, Anil, "Dense and Accurate Multiview Stereo Reconstruction Using Differential Geometry of Curves/Surfaces, Lofting and Occlusion Reasoning" (2017). Engineering Theses and Dissertations. Brown Digital Repository. Brown University Library. https://doi.org/10.7301/Z06H4FWN

Relations

Collection:

Engineering Theses and Dissertations

Theses and Dissertations for the Engineering department.
...