Leveraging photogrammetric mesh models for aerial-ground feature point matching toward integrated 3D reconstruction

Zhu, Qing; Wang, Zhendong; Han, Hu; Xie, Linfu; Ge, Xuming; Zhang, Yeting

doi:10.1016/j.isprsjprs.2020.05.024

Cited by 37 publications

(32 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Even though we have successfully tested the algorithm on simulated drones, we still reduced the drone motion to the plane (see a video here: https://youtu.be/A6A7--rLkfo). While the proposed algorithm itself can work both in 2D and 3D and several techniques can be used for 3D mapping [29,30], defining h-signatures and getting accurate tether configuration data becomes much more complex in 3D. Further research is required in this aspect.…”

Section: Discussionmentioning

confidence: 99%

Tangle-Free Exploration with a Tethered Mobile Robot

Shapovalov

Pereira

2020

Remote Sensing

View full text Add to dashboard Cite

Exploration and remote sensing with mobile robots is a well known field of research, but current solutions cannot be directly applied for tethered robots. In some applications, tethers may be very important to provide power or allow communication with the robot. This paper presents an exploration algorithm that guarantees complete exploration of arbitrary environments within the length constraint of the tether, while keeping the tether tangle-free at all times. While we also propose a generalized algorithm that can be used with several exploration strategies, our implementation uses a modified frontier-based exploration approach, where the robot chooses its next goal in the frontier between explored and unexplored regions of the environment. The basic idea of the algorithm is to keep an estimate of the tether configuration, including length and homotopy, and decide the next robot path based on the difference between the current tether length and the shortest tether length at the next goal position. Our algorithm is provable correct and was tested and evaluated using both simulations and real-world experiments.

show abstract

Section: Discussionmentioning

confidence: 99%

Tangle-Free Exploration with a Tethered Mobile Robot

Shapovalov

Pereira

2020

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…11. The image pair P1-P4 is from BF [27], Centre of Dortmund [51], Zeche of Zurich [51], and SWJTU-BLD [4], respectively. In Fig.…”

Section: A Datasetsmentioning

confidence: 99%

“…While aerial view images provide a larger and more complete range of top views, street view images delineate points of interest and details of urban façades. Establishing correspondence between street (ground) and aerial view images can facilitate a variety of urban applications that require supplementary information from both views, ranging from geolocating [1], [2], urban 3D modeling [3], [4], [5], and autonomous driving [6] to ancient architecture preservation [7]. However, establishing such correspondence is a particularly challenging problem due to the drastic changes in viewpoint, scale, and illumination between ground and aerial images.…”

Section: Introductionmentioning

confidence: 99%

“…It is therefore particularly hard for existing mismatch removal methods to create a separability constraint for such a putative match set. A different way of matching local features between ground and aerial images is to warp one view into another by using camera pose and 3D geometric information [3], [4], [10], and performing feature matching on synthetic images. Some of these methods have been successfully applied in practical applications.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Lattice-Point Mutually Guided Ground-to-Aerial Feature Matching for Urban Scene Images

Zheng

Xiong

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Ground-to-aerial feature matching bridges information from cross-view images, which enables optimized urban applications, e.g., pixel-level geolocating and complete urban 3D reconstruction. However, urban ground and aerial images typically suffer from drastic changes in viewpoint, scale, and illumination, together with repetitive patterns. Thus, direct matching of local features between ground and aerial images is particularly difficult because of the low similarity of local descriptors and high ambiguity in true-false match discrimination. For this challenging task, we propose a novel lattice-point mutually guided matching (LPMG) method in this paper. We specifically address two key issues: 1) reducing descriptor variance and 2) enhancing true-false match discriminability. The former is solved by recovering the geometry and appearance of the underlying image region in 3D through automatic view rectification on ground and aerial images. The latter is circumvented by replacing the conventional mismatch removal with an LPMG strategy. In this strategy, the topology structure of repeated façade elements (i.e., lattice), and the high reliable point matching seeds, are first extracted from the rectified ground and aerial images. Then, the point matching seeds guide the self-similar lattice tiles from two views to be precisely aligned, thereby estimating an accurate transformation model from lattice tile correspondences. Finally, the estimated model powerfully supervises the differentiation of true and false matches from the entire putative match set. Extensive experiments conducted on several datasets show that our method can obtain a considerable number of nearly pure correct matches from urban ground and aerial images, significantly outperforming those existing methods.

show abstract

“…Hence, the SLAM technique is believed to enhance the positioning ability. Visual SLAM is a major stream in the research field, including stereo vision approaches [ 30 ], stereo with wide baseline approaches [ 31 ] as well as localization precision improvement in GNSS-denied areas [ 32 ]. Different from visual SLAM, Lidar SLAM [ 33 , 34 , 35 ] is implemented to compensate and gain better position reliability in open roads, where the features in urban environments can be sufficiently extracted.…”

Section: Related Workmentioning

confidence: 99%

Cost Effective Mobile Mapping System for Color Point Cloud Reconstruction

Cheng

Hsu

Wang

2020

Sensors

View full text Add to dashboard Cite

Survey-grade Lidar brands have commercialized Lidar-based mobile mapping systems (MMSs) for several years now. With this high-end equipment, the high-level accuracy quality of point clouds can be ensured, but unfortunately, their high cost has prevented practical implementation in autonomous driving from being affordable. As an attempt to solve this problem, we present a cost-effective MMS to generate an accurate 3D color point cloud for autonomous vehicles. Among the major processes for color point cloud reconstruction, we first synchronize the timestamps of each sensor. The calibration process between camera and Lidar is developed to obtain the translation and rotation matrices, based on which color attributes can be composed into the corresponding Lidar points. We also employ control points to adjust the point cloud for fine tuning the absolute position. To overcome the limitation of Global Navigation Satellite System/Inertial Measurement Unit (GNSS/IMU) positioning system, we utilize Normal Distribution Transform (NDT) localization to refine the trajectory to solve the multi-scan dispersion issue. Experimental results show that the color point cloud reconstructed by the proposed MMS has a position error in centimeter-level accuracy, meeting the requirement of high definition (HD) maps for autonomous driving usage.

show abstract

Leveraging photogrammetric mesh models for aerial-ground feature point matching toward integrated 3D reconstruction

Cited by 37 publications

References 31 publications

Tangle-Free Exploration with a Tethered Mobile Robot

Tangle-Free Exploration with a Tethered Mobile Robot

Lattice-Point Mutually Guided Ground-to-Aerial Feature Matching for Urban Scene Images

Cost Effective Mobile Mapping System for Color Point Cloud Reconstruction

Contact Info

Product

Resources

About