Parallel Structure from Motion for Sparse Point Cloud Generation in Large-Scale Scenes.

Yongtang Bao,Qing Fan,Pengfei Lin,Yao Li,Wenxiang Du,Yue Qi,Zhihui Wang

doi:10.3390/s21113939

Abstract

Scene reconstruction uses images or videos as input to reconstruct a 3D model of a real scene and has important applications in smart cities, surveying and mapping, military, and other fields. Structure from motion (SFM) is a key step in scene reconstruction, which recovers sparse point clouds from image sequences. However, large-scale scenes cannot be reconstructed using a single compute node. Image matching and geometric filtering take up a lot of time in the traditional SFM problem. In this paper, we propose a novel divide-and-conquer framework to solve the distributed SFM problem. First, we use the global navigation satellite system (GNSS) information from images to calculate the GNSS neighborhood. The number of images matched is greatly reduced by matching each image to only valid GNSS neighbors. This way, a robust matching relationship can be obtained. Second, the calculated matching relationship is used as the initial camera graph, which is divided into multiple subgraphs by the clustering algorithm. The local SFM is executed on several computing nodes to register the local cameras. Finally, all of the local camera poses are integrated and optimized to complete the global camera registration. Experiments show that our system can accurately and efficiently solve the structure from motion problem in large-scale scenes.

Highlights

Structure from motion (SFM) has rapidly developed in the field of 3D reconstruction
Considering that the long distance between the images is low or the images have no correlation, we use the global navigation satellite system (GNSS) information from the image to calculate the distance threshold of the GNSS effective neighbors and perform image matching for each image with its effective neighbors
We propose a method for calculating the GNSS neighborhood, which greatly reduces the time of image matching and ensures the robustness of the matching relationship

Summary

Introduction

Structure from motion (SFM) has rapidly developed in the field of 3D reconstruction. Image feature extraction and matching have generally achieved considerable success in computer vision. In the entire SFM step, image matching and epipolar constraint filtering are the most time consuming. To solve the problem of time consumption, Li et al [1] used the principle of spatial angular order to improve efficiency, which assumes that angular order of neighboring points relating to one correspondence remains invariant under a variety of transformation. This constraint was used to remove outliers from initial matches of image pairs [2,3]. The original squared time complexity of image matching is optimized for linear complexity

Results

Discussion

Conclusion