Online Visual Place Recognition via Saliency Re-identification

Wang, Han; Wang, Chen; Xie, Lihua

doi:10.1109/iros45743.2020.9341703

Cited by 17 publications

(12 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The works in the first category focus on the localization on fine granularity scales, such as landmarks [3,5,46,49] or at city-scale granularity [1,17,25,39,44]. In general, the solutions that are employed for such problems are based on retrieval systems that match the query images with ones from a background collection and then apply a post-processing scheme to estimate the final location.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation

Kordopatis-Zilos

Galopoulos

Papadopoulos

et al. 2021

Proceedings of the 2021 International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

In this paper, we address the problem of global-scale image geolocation, proposing a mixed classification-retrieval scheme. Unlike other methods that strictly tackle the problem as a classification or retrieval task, we combine the two practices in a unified solution leveraging the advantages of each approach with two different modules. The first leverages the EfficientNet architecture to assign images to a specific geographic cell in a robust way. The second introduces a new residual architecture that is trained with contrastive learning to map input images to an embedding space that minimizes the pairwise geodesic distance of same-location images. For the final location estimation, the two modules are combined with a search-within-cell scheme, where the locations of most similar images from the predicted geographic cell are aggregated based on a spatial clustering scheme. Our approach demonstrates very competitive performance on four public datasets, achieving new state-of-the-art performance in fine granularity scales, i.e., 15.0% at 1km range on Im2GPS3k. CCS CONCEPTS• Computing methodologies → Computer vision problems;• Information systems → Geographic information systems.

show abstract

Section: Related Workmentioning

confidence: 99%

“…a particular kind of scenes, e.g., landmarks [3,5,46], to be from a particular area [1,17,44], or on whether we are using different inputs, e.g., a sequence of images per scene [2,31,32], or aerial imagery [24,34,43]. In this study, we focus on global-scale location estimation from single images, which is the most challenging problem setting.…”

mentioning

confidence: 99%

Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation

Kordopatis-Zilos

Galopoulos

Papadopoulos

et al. 2021

Proceedings of the 2021 International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

show abstract

“…In [45], an unsupervised semi-semantic auto-encoder model DeepLab_AE was designed to obtain semantic features of scenes. The authors of [46] proposed a place recognition method via re-identification of salient objects. Oh et al [47] performed loop detection by matching graphs of detected objects.…”

Section: Semantic-information-based Loop Closure Detectionmentioning

confidence: 99%

“…DBoW2 [34], OpenFABMAP [33], SRLCD [46], and BoWT-LCD [39] were selected to complete comparative experiments with SVG-Loop. DBoW2 and OpenFABMAP are the most popular and practical visual-based methods in loop closure detection.…”

Section: Outdoor Datasetmentioning

confidence: 99%

SVG-Loop: Semantic–Visual–Geometric Information-Based Loop Closure Detection

Yuan

Zhou

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Loop closure detection is an important component of visual simultaneous localization and mapping (SLAM). However, most existing loop closure detection methods are vulnerable to complex environments and use limited information from images. As higher-level image information and multi-information fusion can improve the robustness of place recognition, a semantic–visual–geometric information-based loop closure detection algorithm (SVG-Loop) is proposed in this paper. In detail, to reduce the interference of dynamic features, a semantic bag-of-words model was firstly constructed by connecting visual features with semantic labels. Secondly, in order to improve detection robustness in different scenes, a semantic landmark vector model was designed by encoding the geometric relationship of the semantic graph. Finally, semantic, visual, and geometric information was integrated by fuse calculation of the two modules. Compared with art-of-the-state methods, experiments on the TUM RBG-D dataset, KITTI odometry dataset, and practical environment show that SVG-Loop has advantages in complex environments with varying light, changeable weather, and dynamic interference.

show abstract

“…For example, in autonomous exploration, an efficient and robust object encoding benefits the decision process when a robot revisits a specific landmark object [1]. Without the capability of object re-identification, a semantic SLAM system may easily drift and subsequently leads to an unreliable localization [2]. However, existing object encoding methods easily produce false matches due to viewpoint or scaling changes, hence a robust and efficient object encoding method is necessary for many robotic applications.…”

Section: Introductionmentioning

confidence: 99%

AirCode: A Robust Object Encoding Method

Xu,

Wang,

Chen

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Object encoding and identification is crucial for many robotic tasks such as autonomous exploration and semantic relocalization. Existing works heavily rely on the tracking of detected objects but difficult to recall revisited objects precisely. In this paper, we propose a novel object encoding method based on a graph of key-points. To be robust to the number of keypoints detected, we propose a feature sparse encoding and object dense encoding method to ensure that each key-point can only affect a small part of the object descriptors, leading it robust to viewpoint changes, scaling, occlusion, and even object deformation. In the experiments, we show that it achieves superior performance for object identification than the stateof-the art algorithm and is able to provide reliable semantic relocalization. It is a plug-and-play module and we expect that it will play an important role in the robotic applications.

show abstract

Online Visual Place Recognition via Saliency Re-identification

Cited by 17 publications

References 34 publications

Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation

Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation

SVG-Loop: Semantic–Visual–Geometric Information-Based Loop Closure Detection

AirCode: A Robust Object Encoding Method

Contact Info

Product

Resources

About