Title:
Rigid Partitioning Techniques for Efficiently Generating 3D Reconstructions from Images

Simple item page
dc.contributor.advisor Essa, Irfan
dc.contributor.author Steedly, Drew en_US
dc.contributor.committeeMember Mynatt, Elizabeth D.
dc.contributor.committeeMember Bell, Genevieve
dc.contributor.committeeMember Foley, James
dc.contributor.committeeMember Forlizzi, Jodi
dc.contributor.department Computing en_US
dc.date.accessioned 2005-03-01T19:44:31Z
dc.date.available 2005-03-01T19:44:31Z
dc.date.issued 2004-12-01 en_US
dc.description.abstract This thesis explores efficient techniques for generating 3D reconstructions from imagery. Non-linear optimization is one of the core techniques used when computing a reconstruction and is a computational bottleneck for large sets of images. Since non-linear optimization requires a good initialization to avoid getting stuck in local minima, robust systems for generating reconstructions from images build up the reconstruction incrementally. A hierarchical approach is to split up the images into small subsets, reconstruct each subset independently and then hierarchically merge the subsets. Rigidly locking together portions of the reconstructions reduces the number of parameters needed to represent them when merging, thereby lowering the computational cost of the optimization. We present two techniques that involve optimizing with parts of the reconstruction rigidly locked together. In the first, we start by rigidly grouping the cameras and scene features from each of the reconstructions being merged into separate groups. Cameras and scene features are then incrementally unlocked and optimized until the reconstruction is close to the minimum energy. This technique is most effective when the influence of the new measurements is restricted to a small set of parameters. Measurements that stitch together weakly coupled portions of the reconstruction, though, tend to cause deformations in the low error modes of the reconstruction and cannot be efficiently incorporated with the previous technique. To address this, we present a spectral technique for clustering the tightly coupled portions of a reconstruction into rigid groups. Reconstructions partitioned in this manner can closely mimic the poorly conditioned, low error modes, and therefore efficiently incorporate measurements that stitch together weakly coupled portions of the reconstruction. We explain how this technique can be used to scalably and efficiently generate reconstructions from large sets of images. en_US
dc.description.degree Ph.D. en_US
dc.format.extent 2791483 bytes
dc.format.mimetype application/pdf
dc.identifier.uri http://hdl.handle.net/1853/4925
dc.language.iso en_US
dc.publisher Georgia Institute of Technology en_US
dc.subject Bundle adjustment en_US
dc.subject Photogrammetry
dc.subject Structure from motion
dc.subject Computer vision
dc.subject Spectral partitioning
dc.subject Sparse matrix
dc.subject.lcsh Three-dimensional imaging en_US
dc.subject.lcsh Photogrammetry en_US
dc.subject.lcsh Image reconstruction en_US
dc.title Rigid Partitioning Techniques for Efficiently Generating 3D Reconstructions from Images en_US
dc.type Text
dc.type.genre Dissertation
dspace.entity.type Publication
local.contributor.advisor Essa, Irfan
local.contributor.corporatename College of Computing
relation.isAdvisorOfPublication 84ae0044-6f5b-4733-8388-4f6427a0f817
relation.isOrgUnitOfPublication c8892b3c-8db6-4b7b-a33a-1b67f7db2021
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
steedly_drew_e_200412_phd.pdf
Size:
2.66 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Theses and Dissertations