VIO-Aided Structure from Motion Under Challenging Environments

Autor:	Jiang, Zijie, Taira, Hajime, Miyashita, Naoyuki, Okutomi, Masatoshi
Rok vydání:	2021
Předmět:	Computer Science - Computer Vision and Pattern Recognition
Druh dokumentu:	Working Paper
Popis:	In this paper, we present a robust and efficient Structure from Motion pipeline for accurate 3D reconstruction under challenging environments by leveraging the camera pose information from a visual-inertial odometry. Specifically, we propose a geometric verification method to filter out mismatches by considering the prior geometric configuration of candidate image pairs. Furthermore, we introduce an efficient and scalable reconstruction approach that relies on batched image registration and robust bundle adjustment, both leveraging the reliable local odometry estimation. Extensive experimental results show that our pipeline performs better than the state-of-the-art SfM approaches in terms of reconstruction accuracy and robustness for challenging sequential image collections. Comment: This manuscript was accepted and presented in the 22th IEEE International Conference on Industrial Technology (ICIT2021)
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2101.09657 Zobrazit plný text záznamu View this record from Arxiv