A fast and accurate bundle adjustment method for very large-scale data. (September 2020)

• Record Type: Journal Article
• Authors: Zheng, Maoteng; Zhang, Fayong; Zhu, Junfeng; Zuo, Zejun
• Is Part Of: Computers & geosciences. Volume 142(2020)
• Publication Date: 2020-09
• Subjects: Bundle adjustment -- Large-scale data -- Preconditioned conjugate gradient -- Sparse matrix compression; Environmental policy -- Periodicals; 550.5
• DOI: 10.1016/j.cageo.2020.104539
• Languages: English
• ISSNs: 0098-3004

Abstract

Abstract: Bundle adjustment with very large scale datasets has drew much concern recently in both photogrammetry and computer vision communities. Different from the existing out-of-core and distributed methods for large scale datasets, we propose a fast and accurate bundle adjustment method which still uses the framework of the traditional Levenberg Marquardt (LM) algorithm while adopting preconditioned conjugate gradient (PCG) to iteratively solve normal equation, and using point resampling scheme and normal matrix compression to decrease the memory requirement and computational complexity. Preliminary results show that our method running on a single laptop computer with i7 2.6 GHz CPU and 8 GB RAM is even faster than the state-of-the-art distributed method deployed on a large distributed computer system with multiple computers each of which is equipped with CPU i7-4770K 3.5 GHz with 8 threads and 32 GB RAM and connected with each other at the speed of 10 MB/s. The proposed method is also more accurate. Highlights: A fast and accurate bundle adjustment method is introduced. The proposed method applies point resampling scheme to improve the efficiency. An advanced matrix compression format is designed to decrease the memory requirement.