Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Sign Up
Cited by 2,584

The DOI matches its Crossref record; 23 of the 23 references it cites were found in a registry or an index.

NetVLAD: CNN Architecture for Weakly Supervised Place Recognition

Open access
Arandjelovic, Relja, Gronat, Petr, Torii, Akihiko, Pajdla, Tomas, Sivic, Josef·2016·2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
DOI 10.1109/cvpr.2016.572
Open this paper in CoChat — free account

Abstract

We tackle the problem of large scale visual place recognition, where the task is to quickly and accurately recognize the location of a given query photograph. We present the following four principal contributions. First, we develop a convolutional neural network (CNN) architecture that is trainable in an end-to-end manner directly for the place recognition task. The main component of this architecture, NetVLAD, is a new generalized VLAD layer, inspired by the "Vector of Locally Aggregated Descriptors" image representation commonly used in image retrieval. The layer is readily pluggable into any CNN architecture and amenable to training via backpropagation. Second, we create a new weakly supervised ranking loss, which enables end-to-end learning of the architecture's parameters from images depicting the same places over time downloaded from Google Street View Time Machine. Third, we develop an efficient training procedure which can be applied on very large-scale weakly labelled tasks. Finally, we show that the proposed architecture and training procedure significantly outperform non-learnt image representations and off-the-shelf CNN descriptors on challenging place recognition and image retrieval benchmarks.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/cvpr.2016.572

Cites

23 of 23 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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

Show all 23 references

Research works printed but not linked: 73

  1. simonyan, 2015, Very deep convolutional networks for large-scale image recognition, Proc ICLR
  2. simonyan, 2013, Deep Fisher networks for large-scale image classification, NIPS
  3. 10.1007/978-3-642-33718-5_18 · DOI
  4. 10.1109/ICCV.2013.177 · DOI
  5. 10.15607/RSS.2015.XI.022 · DOI · Paper page
  6. 10.1109/CVPR.2014.182 · DOI
  7. 10.1016/j.patcog.2014.04.007 · DOI
  8. 10.1109/CVPR.2015.7298790 · DOI · Paper page
  9. 10.1109/CVPR.2014.417 · DOI
  10. 10.1109/TPAMI.2011.235 · DOI · Paper page
  11. 10.1109/CVPR.2007.382970 · DOI
  12. kotzias, 2014, Deep multi-instance transfer learning, CoRR, abs 1411 3128
  13. knopp, 2010, Avoiding confusing features in place recognition, Proc ECCV
  14. 10.1109/CVPRW.2014.131 · DOI · Paper page
  15. 10.1109/ICCV.2015.243 · DOI · Paper page
  16. razavian, 2015, A baseline for visual instance retrieval with deep convolutional networks, Proc ICLR
  17. 10.1109/ICCV.2011.6126302 · DOI · Paper page
  18. 10.5244/C.26.76 · DOI · Paper page
  19. 10.1109/CVPR.2009.5206609 · DOI · Paper page
  20. 10.1109/CVPR.2007.383150 · DOI · Paper page
  21. 10.1109/CVPR.2010.5540039 · DOI · Paper page
  22. schultz, 2004, Learning a distance metric from relative comparisons, NIPS
  23. sermanet, 2013, OverFeat: Integrated recognition, localization and detection using convolutional networks, CoRR, abs 1312 6229
  24. simo-serra, 2014, Fracking deep convolutional image descriptors, CoRR, abs 1412 6537
  25. 0, Supplementary material (appendix) for the paper
  26. 0, Project webpage (code/networks)
  27. 10.1109/CVPR.2014.81 · DOI · Paper page
  28. 10.1017/S026988890999035X · DOI · Paper page
  29. 10.1109/CVPR.2013.122 · DOI
  30. gong, 2014, Multi-scale orderless pooling of deep convolutional activation features, Proc ECCV
  31. jégou, 2012, Negative evidences and co-occurrences in image retrieval: the benefit of PCA and whitening, Proc ECCV
  32. 10.1109/ICCV.2015.19 · DOI
  33. 10.1109/CVPR.2007.383266 · DOI · Paper page
  34. 10.1109/CVPR.2013.211 · DOI
  35. 10.1109/CVPR.2011.5995373 · DOI · Paper page
  36. qin, 2014, Learning to rank bag-of-word histograms for large-scale object retrieval, Proc BMVC
  37. qin, 2014, Quantized kernel learning for feature matching, NIPS
  38. philbin, 2010, Descriptor learning for efficient retrieval, Proc ECCV
  39. 10.1109/CVPR.2008.4587635 · DOI · Paper page
  40. 10.1109/CVPR.2007.383172 · DOI · Paper page
  41. 10.1109/CVPR.2010.5540009 · DOI · Paper page
  42. 10.1109/CVPR.2013.96 · DOI
  43. 10.1109/CVPR.2011.5995610 · DOI · Paper page
  44. li, 2012, Worldwide pose estimation using 3D point clouds, Proc ECCV
  45. 10.1109/CVPR.2011.5995601 · DOI · Paper page
  46. 10.1109/ICCV.2007.4408891 · DOI · Paper page
  47. 10.1109/CVPR.2015.7299007 · DOI · Paper page
  48. csurka, 2004, Visual categorization with bags of keypoints, Workshop on Statistical Learning in Computer Vision ECCV, 1
  49. 10.1561/0600000017 · DOI · Paper page
  50. 10.1109/ICCVW.2009.5457541 · DOI
  51. 10.15607/RSS.2009.V.039 · DOI · Paper page
  52. 10.1109/CVPR.2014.180 · DOI · Paper page
  53. 10.1145/2502081.2502171 · DOI
  54. viola, 2005, Multiple instance boosting for object detection, NIPS
  55. 10.1109/CVPR.2013.119 · DOI · Paper page
  56. zhou, 2014, Learning deep features for scene recognition using places database, NIPS
  57. arandjelovi?, 2013, All about VLAD, Proc CVPR
  58. arandjelovi?, 2012, Three things everyone should know to improve object retrieval, Proc CVPR
  59. aubry, 2014, Painting-to-3D model alignment via discriminative visual elements, ACM Transactions on Graphics (TOG), 33, 14, 10.1145/2591009 · DOI
  60. arandjelovi?, 2014, DisLocation: Scalable descriptor distinctiveness for location recognition, Proc ACCV
  61. weinberger, 2006, Distance metric learning for large margin nearest neighbor classification, NIPS
  62. babenko, 2015, Aggregating local deep features for image retrieval, Proc ICCV
  63. 10.1109/CVPR.2009.5206839 · DOI · Paper page
  64. zeiler, 2014, Visualizing and understanding convolutional networks, Proc ECCV
  65. 10.1109/CVPR.2015.7298924 · DOI
  66. babenko, 2014, Neural codes for image retrieval, Proc ECCV
  67. 10.1109/ICRA.2014.6906961 · DOI
  68. 10.1007/978-3-642-15555-0_23 · DOI
  69. mikulik, 2010, Learning a fine vocabulary, Proc ECCV
  70. middelberg, 2014, Scalable 6-DOF localization on mobile devices, Proc ECCV
  71. 10.1109/ICCV.2015.170 · DOI · Paper page
  72. 10.1109/CVPR.2015.7299135 · DOI · Paper page
  73. maddern, 2014, Towards robust night and day place recognition using visible and thermal imaging, Proc Intl Conf on Robotics and Automation

Cited by

See all 2,584

Metadata

Authors
Arandjelovic, R., Gronat, P., Torii, A., Pajdla, T., & Sivic, J.
Venue
2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2016-06
Volume
40
Issue
6
Pages
5297-5307
ISSN
1063-6919 (ISSN-L)
DOI
10.1109/cvpr.2016.572
OpenAlex
W2179042386
Cited by
2,584
References
23
FWCI
60.38
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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: NetVLAD: CNN Architecture for Weakly Supervised Place Recognition. https://app.cochat.ai/papers/10.1109/cvpr.2016.572
The paper
Arandjelovic, R., Gronat, P., Torii, A., Pajdla, T., & Sivic, J. (2016). NetVLAD: CNN Architecture for Weakly Supervised Place Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). https://doi.org/10.1109/cvpr.2016.572

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.