View Selection with Geometric Uncertainty Modelling

Cheng Peng, Volkan Isler

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations


Estimating positions of world points from features observed in images is a key problem in 3D reconstruction, image mosaicking, simultaneous localization and mapping and structure from motion. We consider a special instance in which there is a dominant ground plane G viewed from a parallel viewing plane S above it. Such instances commonly arise, for example, in aerial photography. Consider a world point g ∈ G and its worst case reconstruction uncertainty ε(g, S) obtained by merging all possible views of g chosen from S. We first show that one can pick two views sp and sq such that the uncertainty ε(g, {sp, sq}) obtained using only these two views is almost as good as (i.e, within a small constant factor of) ε(g, S). Next, we extend the result to the entire ground plane G and show that one can pick a small subset of S ⊆ S (which grows only linearly with the area of G) and still obtain a constant factor approximation, for every point g ∈ G, to the minimum worst case estimate obtained by merging all views in S. Finally, we present a multi-resolution view selection method which extends our techniques to non-planar scenes. We show that the method can produce rich and accurate dense reconstructions with a small number of views. Our results provide a view selection mechanism with provable performance guarantees which can drastically increase the speed of scene reconstruction algorithms. In addition to theoretical results, we demonstrate their effectiveness in an application where aerial imagery is used for monitoring farms and orchards.

Original languageEnglish (US)
Title of host publicationRobotics
Subtitle of host publicationScience and Systems XIV
EditorsHadas Kress-Gazit, Siddhartha S. Srinivasa, Tom Howard, Nikolay Atanasov
PublisherMIT Press Journals
ISBN (Print)9780992374747
StatePublished - 2018
Event14th Robotics: Science and Systems, RSS 2018 - Pittsburgh, United States
Duration: Jun 26 2018Jun 30 2018

Publication series

NameRobotics: Science and Systems
ISSN (Electronic)2330-765X


Conference14th Robotics: Science and Systems, RSS 2018
Country/TerritoryUnited States

Bibliographical note

Funding Information:
We would like to acknowledge the supports by a MN State LCCMR grant and NSF Awards 1525045 and 1617718.

Publisher Copyright:
© 2018, MIT Press Journals. All rights reserved.


Dive into the research topics of 'View Selection with Geometric Uncertainty Modelling'. Together they form a unique fingerprint.

Cite this