Rethinking Visual Geo-localization for Large-Scale Applications

Gabriele, Berton,; Masone, Carlo; Caputo, Barbara

doi:10.1109/cvpr52688.2022.00483

Cited by 69 publications

(54 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We can see that the Spectral approach selects candidates in different regions of the pose graph to better increase the estimation accuracy, while the Greedy approach tends to select redundant candidates in high similarity regions. global descriptors of lidar scans and the recent CNN-based CosPlace [33] for images. We use nearest neighbors based on cosine similarity for descriptor matching.…”

Section: A Global Matchingmentioning

confidence: 99%

Swarm-SLAM : Sparse Decentralized Collaborative Simultaneous Localization and Mapping Framework for Multi-Robot Systems

Lajoie¹,

Beltrame²

2023

Preprint

View full text Add to dashboard Cite

Collaborative Simultaneous Localization And Mapping (C-SLAM) is a vital component for successful multirobot operations in environments without an external positioning system, such as indoors, underground or underwater. In this paper, we introduce Swarm-SLAM, an open-source C-SLAM system that is designed to be scalable, flexible, decentralized, and sparse, which are all key properties in swarm robotics. Our system supports inertial, lidar, stereo, and RGB-D sensing, and it includes a novel inter-robot loop closure prioritization technique that reduces communication and accelerates convergence. We evaluated our ROS-2 implementation on five different datasets, and in a real-world experiment with three robots communicating through an ad-hoc network.

show abstract

Section: A Global Matchingmentioning

confidence: 99%

Swarm-SLAM : Sparse Decentralized Collaborative Simultaneous Localization and Mapping Framework for Multi-Robot Systems

Lajoie¹,

Beltrame²

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The calculation of this global description is an active topic of research, and related algorithms are capable of achieving unprecedented performance. For instance, Hloc [22] learns to simultaneously predict global and local features or CosPlace [23] provides a learned global descriptor without an intermediate step on local feature aggregation. We will make use of state-of-the-art global image descriptors for VPR as a tool.…”

Section: Related Workmentioning

confidence: 99%

“…The following methods that show top performance on popular VPR datasets are considered: NetVLAD [15], Cos-Place [23], SuperGlue-based approach [37] and a combination of NetVLAD and SuperGlue. The SuperGlue approach counts the number of keypoint matches between a query image and each image in the database and the image with maximum amount of matches is taken as the result of image retrieval.…”

Section: Visual Place Recognitionmentioning

confidence: 99%

Dominating Set Database Selection for Visual Place Recognition

Kornilova¹,

Moskalenko²,

Pushkin³

et al. 2023

Preprint

View full text Add to dashboard Cite

“…Recently, the advents of FILD++ [59], HEAPUtil [60] and CosPlace [61] has facilitated rapid advancements in VPR domain.…”

Section: Related Workmentioning

confidence: 99%

CoCALC: A Self-Supervised Visual Place Recognition Approach Combining Appearance and Geometric Information

Li¹,

Wang

Shi³

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Visual place recognition (VPR) is considered among the most complicated tasks in SLAM due to the multiple challenges of drastic variations in both appearance and viewpoint. To address this issue, this article presents a self-supervised and lightweight VPR approach (namely CoCALC) that fully utilizes the appearance and geometric information provided by images. The main thing that makes CoCALC ultralightweight (only 0.27 MB) is our use of Depthwise Separable Convolution (DSC), a simple but effective architecture that enables our model to generate a more robust image representation. The network trained specifically for VPR can efficiently extract deep convolutional features from salient image regions that have relatively higher entropy, thereby expanding its applications on resource-limited platforms without GPUs. To further eliminate the negative consequences of the high percent false matches, a novel band-matrixbased geometric check is employed to filter out the incorrect matching of image patches, and the impact of different bandwidths on the recall rate is discussed. Results on several benchmark datasets confirm that the proposed CoCALC can yield state-of-the-art performance and superior generalization with acceptable efficiency. All relevant codes are provided at https://github.com/LiKangyuLKY/CoCALC-VPR for further studies.

show abstract

Rethinking Visual Geo-localization for Large-Scale Applications

Cited by 69 publications

References 40 publications

Swarm-SLAM : Sparse Decentralized Collaborative Simultaneous Localization and Mapping Framework for Multi-Robot Systems

Swarm-SLAM : Sparse Decentralized Collaborative Simultaneous Localization and Mapping Framework for Multi-Robot Systems

Dominating Set Database Selection for Visual Place Recognition

CoCALC: A Self-Supervised Visual Place Recognition Approach Combining Appearance and Geometric Information

Contact Info

Product

Resources

About