TransVPR: Transformer-Based Place Recognition with Multi-Level Attention Aggregation
Open accessThe DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.
Abstract
Visual place recognition is a challenging task for applications such as autonomous driving navigation and mobile robot localization. Distracting elements presenting in complex scenes often lead to deviations in the perception of visual place. To address this problem, it is crucial to integrate information from only task-relevant regions into image representations. In this paper, we introduce a novel holistic place recognition model, TransVPR, based on vision Transformers. It benefits from the desirable property of the self-attention operation in Transformers which can naturally aggregate task-relevant features. Attentions from multiple levels of the Transformer, which focus on different regions of interest, are further combined to generate a global image representation. In addition, the output tokens from Transformer layers filtered by the fused attention mask are considered as key-patch descriptors, which are used to perform spatial matching to re-rank the candidates retrieved by the global image features. The whole model allows end-to-end training with a single objective and image-level supervision. TransVPR achieves state-of-the-art performance on several real-world benchmarks while maintaining low computational time and storage requirements.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
Fischler, M. A., & Bolles, R. C. (1981) · Communications of the ACM · doi:10.1145/358669.358692 · Cited by 25,893
End-to-End Object Detection with Transformers
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., & Zagoruyko, S. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58452-8_13 · Cited by 18,285
End-to-End Object Detection with Transformers
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., & Zagoruyko, S. (2020) · arXiv · doi:10.48550/arxiv.2005.12872 · Cited by 18,253
Object recognition from local scale-invariant features
Lowe, D. G. (1999) · Proceedings of the Seventh IEEE International Conference on Computer Vision · doi:10.1109/iccv.1999.790410 · Cited by 16,472 · cited 2×
“In early VPR systems [1, 11, 25, 30, 31, 46], traditional methods such as SIFT [26], SURF [4] and ORB [39] have been widely used to represent a small patch centered around a detected keypoint.”
MethodFaceNet: A unified embedding for face recognition and clustering
Schroff, F., Kalenichenko, D., & Philbin, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298682 · Cited by 15,267
SURF: Speeded Up Robust Features
Bay, H., Tuytelaars, T., & Van Gool, L. (2006) · Lecture Notes in Computer Science · doi:10.1007/11744023_32 · Cited by 14,955
ORB: An efficient alternative to SIFT or SURF
Rublee, E., Rabaud, V., Konolige, K., & Bradski, G. (2011) · 2011 International Conference on Computer Vision · doi:10.1109/iccv.2011.6126544 · Cited by 11,037
ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras
Mur-Artal, R., & Tardos, J. D. (2017) · IEEE Transactions on Robotics · doi:10.1109/tro.2017.2705103 · Cited by 6,898 · cited 1×
“In early VPR systems [1, 11, 25, 30, 31, 46], traditional methods such as SIFT [26], SURF [4] and ORB [39] have been widely used to represent a small patch centered around a detected keypoint.”
MethodVideo Google: a text retrieval approach to object matching in videos
Sivic, & Zisserman (2003) · Proceedings Ninth IEEE International Conference on Computer Vision · doi:10.1109/iccv.2003.1238663 · Cited by 6,367 · cited 1×
“Some traditional techniques, such as Bag of Words (BoW) [10, 49], Fisher Kernel [21, 35, 36], and Vector of Locally Aggregated Descriptors (VLAD) [3,20], have been used to assign visual words to images.”
MethodPlaces: A 10 Million Image Database for Scene Recognition
Zhou, B., Lapedriza, A., Khosla, A., Oliva, A., & Torralba, A. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2723009 · Cited by 4,042
1 year, 1000 km: The Oxford RobotCar dataset
Maddern, W., Pascoe, G., Linegar, C., & Newman, P. (2017) · The International Journal of Robotics Research · doi:10.1177/0278364916679498 · Cited by 1,632 · cited 1×
“We evaluated our model on several public benchmark datasets: MSLS [58], Nordland [33, 50], Pitts30k [55] and RobotCar Seasons v2 (RobotCar-S2) [28, 42].”
MethodAggregating Local Image Descriptors into Compact Codes
Jegou, H., Perronnin, F., Douze, M., Sanchez, J., Perez, P., & Schmid, C. (2012) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2011.235 · Cited by 1,503
FAB-MAP: Probabilistic Localization and Mapping in the Space of Appearance
Cummins, M., & Newman, P. (2008) · The International Journal of Robotics Research · doi:10.1177/0278364908090961 · Cited by 1,472 · cited 1×
“In early VPR systems [1, 11, 25, 30, 31, 46], traditional methods such as SIFT [26], SURF [4] and ORB [39] have been widely used to represent a small patch centered around a detected keypoint.”
MethodVisual Place Recognition: A Survey
Lowry, S., Sünderhauf, N., Newman, P., Leonard, J. J., Cox, D., Corke, P., & Milford, M. J. (2016) · IEEE Transactions on Robotics · doi:10.1109/tro.2015.2496823 · Cited by 1,079 · cited 2×
“Visual Place Recognition (VPR) is an essential and challenging problem in autonomous driving and robot localization systems, which is usually defined as an image retrieval problem [27].”
BackgroundParticular object retrieval with integral max-pooling of CNN activations
Tolias, G., Sicre, R., & Jégou, H. (2015) · arXiv · doi:10.48550/arxiv.1511.05879 · Cited by 744 · cited 1×
“Likewise, in deep learning context, some works [2, 29] have incorporated these clustering methods into CNN architectures, while other studies [18,22,37,53] have focused on pooling from CNN feature maps.”
MethodBenchmarking 6DOF Outdoor Visual Localization in Changing Conditions
Sattler, T., Maddern, W., Toft, C., Torii, A., Hammarstrand, L., Stenborg, E., Safari, D., Okutomi, M., Pollefeys, M., Sivic, J., et al. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00897 · Cited by 699 · cited 1×
“We evaluated our model on several public benchmark datasets: MSLS [58], Nordland [33, 50], Pitts30k [55] and RobotCar Seasons v2 (RobotCar-S2) [28, 42].”
MethodLearning With Average Precision: Training Image Retrieval With a Listwise Loss
Revaud, J., Almazan, J., Rezende, R., & Souza, C. D. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00521 · Cited by 408 · cited 1×
“Global image features [2, 8, 18, 21, 37, 38, 54] abstract the whole image into a compact feature vector without geometrical information.”
BackgroundTraining Vision Transformers for Image Retrieval
El-Nouby, A., Neverova, N., Laptev, I., & Jégou, H. (2021) · arXiv · doi:10.48550/arxiv.2102.05644 · Cited by 127 · cited 1×
“[15] introduced vision Transformers to image retrieval tasks by using the [class] token [56] from the final layer as a global feature.”
Method
Cited by
MixVPR: Feature Mixing for Visual Place Recognition
Ali-Bey, A., Chaib-Draa, B., & Giguere, P. (2023) · 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv56688.2023.00301 · Cited by 256
Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition
Wang, C., Chen, S., Song, Y., Xu, R., Zhang, Z., Zhang, J., Yang, H., Zhang, Y., Fu, K., Du, S., et al. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i7.32811 · Cited by 17
“TransVPR (Wang et al. 2022b) leverages the transformer ar-chitecture’s multi-level attention mechanism to derive global features for candidate retrieval, and uses an attention mask to filter the feature map to generate key patch descriptors for reordering candidates.”
Metadata
- Authors
- Wang, R., Shen, Y., Zuo, W., Zhou, S., & Zheng, N.
- Venue
- 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2022-06
- Pages
- 13638-13647
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W4312336574
- Cited by
- 185
- References
- 19
- FWCI
- 15.55
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.