TY - JOUR

T1 - AN EFFICIENT METHOD TO DETECT MUTUAL OVERLAP OF A LARGE SET OF UNORDERED IMAGES FOR STRUCTURE-FROM-MOTION

AU - Wang, X.

AU - Zhan, Z. Q.

AU - Heipke, C.

N1 - Funding Information: The author Xin Wang would like to thank the China Scholarship Council (CSC) for financially supporting his PhD study at Leibniz Universität Hannover, Germany. Publisher Copyright: © 2017 Copernicus GmbH. All rights reserved. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 2017/5/30

Y1 - 2017/5/30

N2 - Recently, low-cost 3D reconstruction based on images has become a popular focus of photogrammetry and computer vision research. Methods which can handle an arbitrary geometric setup of a large number of unordered and convergent images are of particular interest. However, determining the mutual overlap poses a considerable challenge. We propose a new method which was inspired by and improves upon methods employing random k-d forests for this task. Specifically, we first derive features from the images and then a random k-d forest is used to find the nearest neighbours in feature space. Subsequently, the degree of similarity between individual images, the image overlaps and thus images belonging to a common block are calculated as input to a structure-from-motion (sfm) pipeline. In our experiments we show the general applicability of the new method and compare it with other methods by analyzing the time efficiency. Orientations and 3D reconstructions were successfully conducted with our overlap graphs by sfm. The results show a speed-up of a factor of 80 compared to conventional pairwise matching, and of 8 and 2 compared to the VocMatch approach using 1 and 4 CPU, respectively.

AB - Recently, low-cost 3D reconstruction based on images has become a popular focus of photogrammetry and computer vision research. Methods which can handle an arbitrary geometric setup of a large number of unordered and convergent images are of particular interest. However, determining the mutual overlap poses a considerable challenge. We propose a new method which was inspired by and improves upon methods employing random k-d forests for this task. Specifically, we first derive features from the images and then a random k-d forest is used to find the nearest neighbours in feature space. Subsequently, the degree of similarity between individual images, the image overlaps and thus images belonging to a common block are calculated as input to a structure-from-motion (sfm) pipeline. In our experiments we show the general applicability of the new method and compare it with other methods by analyzing the time efficiency. Orientations and 3D reconstructions were successfully conducted with our overlap graphs by sfm. The results show a speed-up of a factor of 80 compared to conventional pairwise matching, and of 8 and 2 compared to the VocMatch approach using 1 and 4 CPU, respectively.

KW - image orientation

KW - random k-d forest

KW - Unordered set of images

UR - http://www.scopus.com/inward/record.url?scp=85042630242&partnerID=8YFLogxK

U2 - 10.5194/isprs-annals-IV-1-W1-191-2017

DO - 10.5194/isprs-annals-IV-1-W1-191-2017

M3 - Conference article

AN - SCOPUS:85042630242

VL - 4

SP - 191

EP - 198

JO - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

JF - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

SN - 2194-9042

IS - 1/W1

T2 - ISPRS Hannover Workshop 2017 on High-Resolution Earth Imaging for Geospatial Information, HRIGI 2017, City Models, Roads and Traffic , CMRT 2017, Image Sequence Analysis, ISA 2017, European Calibration and Orientation Workshop, EuroCOW 2017

Y2 - 6 June 2017 through 9 June 2017

ER -