Hierarchical Clustering-Aligning Framework Based Fast Large-Scale 3D Reconstruction Using Aerial Imagery

Xiuchuan Xie,Zhi Li,Tao Yang,Yanning Zhang,Dongdong Li

doi:10.3390/rs11030315

Abstract

With extensive applications of Unmanned Aircraft Vehicle (UAV) in the field of remotesensing, 3D reconstruction using aerial images has been a vibrant area of research. However,fast large-scale 3D reconstruction is a challenging task. For aerial image datasets, large scale meansthat the number and resolution of images are enormous, which brings significant computationalcost to the 3D reconstruction, especially in the process of Structure from Motion (SfM). In thispaper, for fast large-scale SfM, we propose a clustering-aligning framework that hierarchicallymerges partial structures to reconstruct the full scene. Through image clustering, an overlappingrelationship between image subsets is established. With the overlapping relationship, we proposea similarity transformation estimation method based on joint camera poses of common images.Finally, we introduce the closed-loop constraint and propose a similarity transformation-based hybridoptimization method to make the merged complete scene seamless. The advantage of the proposedmethod is a significant efficiency improvement without a marginal loss in accuracy. Experimentalresults on the Qinling dataset captured over Qinling mountain covering 57 square kilometersdemonstrate the efficiency and robustness of the proposed method.

Highlights

With the development of Unmanned Aircraft Vehicle (UAV), low-altitude remote sensing is playing an increasingly important role in land cover monitoring [1,2,3], heritage site protection [4], and vegetation observation [5,6,7]
Structure from Motion (SfM) is a 3D reconstruction method used to recover the 3D structure of stationary scenes from a set of projective measurements, via motion estimation of the cameras corresponding to images
In the process of aligning partial reconstructions, we present a robust initial similarity transformation estimation method based on joint camera poses of common images across image subsets without 3D point matches

Summary

Introduction

With the development of Unmanned Aircraft Vehicle (UAV), low-altitude remote sensing is playing an increasingly important role in land cover monitoring [1,2,3], heritage site protection [4], and vegetation observation [5,6,7]. For these applications, aerial imagery-based 3D reconstruction of large-scale scenes is highly desired. Structure from Motion (SfM) is a 3D reconstruction method used to recover the 3D structure of stationary scenes from a set of projective measurements, via motion estimation of the cameras corresponding to images. SfM is the primary task for the research and application of 3D reconstruction using aerial images

Methods

Results

Discussion

Conclusion