Abstract
Unmanned aerial vehicle (UAV) images have become the main remote sensing data sources for varying applications, and structure from motion (SfM) is the golden standard for resuming camera poses. Matching local feature descriptors is the prerequisite for the accurate and complete orientation of UAV images. Recently, some newly proposed learned methods have been shown to outperform the hand-crafted methods, such as the scale invariant feature transform (SIFT) and its variants, and almost all learned methods have been trained and evaluated by using images from the internet with varying focal lengths and varying size. It is of interest to investigate the performance of these learned methods with their pretrained models for feature detection and description in the context of the SfM-based orientation. Thus, this article conducts a comprehensive evaluation of both advanced hand-crafted and newly proposed learned detectors and descriptors by using four UAV datasets. The performance of these selected methods is compared in the context of feature matching and the SfM and (multiview stereo) MVS-based reconstruction. Experimental results demonstrate that the learned descriptors combined with the SIFT-like detectors can provide accurate and complete feature correspondences and achieve better or competitive performance in the SfM and MVS-based reconstruction. For UAV image orientation, the learned descriptors can be an alternative to the existing hand-crafted descriptors without their model retraining. The source codes of this evaluation would be made publicly available.
Highlights
Unmanned aerial vehicles (UAVs) have become a widely used remote sensing platform in the fields of photogrammetry and computer vision [1], [2]
By using four UAV datasets, the performance of these selected methods is compared in the context of feature matching and the structure from motion (SfM) and MVS-based reconstruction
Experimental results demonstrate that learned methods in the separate detector and descriptor group can provide accurate and complete feature correspondences and achieve better or competitive performance in the context of SfM and MVSbased reconstruction compared with hand-crafted methods
Summary
Unmanned aerial vehicles (UAVs) have become a widely used remote sensing platform in the fields of photogrammetry and computer vision [1], [2]. Due to the advantages of low economic costs and flexible data acquisitions, UAVs integrated with non-metric cameras can provide high-resolution and multi-view images for varying applications [3], including transmission line inspection [4], [5], archaeological excavation [6], [7] and agricultural management [8]. Before their applications, the accurate recovery of camera poses is a crucial step in the pipeline of UAV image processing [9]. It is rational and necessary to evaluate their performance in the complete SfM pipeline [24]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have