mlsanityData to 7 Oct 2026

NetVLAD: CNN architecture for weakly supervised place recognition

arXiv 1511.07247 ↗First posted 23 Nov 2015cs.CV cs.LGcited by 935

Not on the map. Why?

Abstract (arXiv)

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 three 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 develop a training procedure, based on a new weakly supervised ranking loss, to learn parameters of the architecture in an end-to-end manner from images depicting the same places over time downloaded from Google Street View Time Machine. Finally, we show that the proposed architecture significantly outperforms non-learnt image representations and off-the-shelf CNN descriptors on two challenging place recognition benchmarks, and improves over current state-of-the-art compact image representations on standard image retrieval benchmarks.

Key facts. Cited by 935 papers in the map (data through 7 Oct 2026). First posted on arXiv 23 Nov 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 141; 2022: 161; 2023: 163; 2024: 167; 2025: 164; 2026: 139.

935 papers.

Sort:

By area:

  1. 2023LightGlue: Local Feature Matching at Light Speedcited by 583
  2. 2021Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognitioncited by 236
  3. 2021CLIP2Video: Mastering Video-Text Retrieval via Image CLIPcited by 223
  4. 2024Global Structure-from-Motion Revisitedcited by 215
  5. 2021Back to the Feature: Learning Robust Camera Localization from Pixels to Posecited by 199
  6. 2024MASt3R-SfM: a Fully-Integrated Solution for Unconstrained Structure-from-Motioncited by 161
  7. 2022Rethinking Visual Geo-localization for Large-Scale Applicationscited by 160
  8. 2023MixVPR: Feature Mixing for Visual Place Recognitioncited by 160
  9. 2023AnyLoc: Towards Universal Visual Place Recognitioncited by 159
  10. 2023Optimal Transport Aggregation for Visual Place Recognitioncited by 151
  11. 2021Human POSEitioning System (HPS): 3D Human Pose Estimation and Self-localization in Large Scenes from Body-Mounted Sensorscited by 137
  12. 2024Efficient LoFTR: Semi-Dense Local Feature Matching with Sparse-Like Speedcited by 119
38 more
  1. 2021Coming Down to Earth: Satellite-to-Street View Synthesis for Geo-Localizationcited by 112
  2. 2022Map-free Visual Relocalization: Metric Pose Relative to a Single Imagecited by 112
  3. 2023EigenPlaces: Training Viewpoint Robust Models for Visual Place Recognitioncited by 109
  4. 2023Sample4Geo: Hard Negative Sampling For Cross-View Geo-Localisationcited by 107
  5. 2023Accelerated Coordinate Encoding: Learning to Relocalize in Minutes using RGB and Posescited by 107
  6. 2024Measuring Style Similarity in Diffusion Modelscited by 106
  7. 2021Novel Visual Category Discovery with Dual Ranking Statistics and Mutual Knowledge Distillationcited by 104
  8. 2021T2VLAD: Global-Local Sequence Alignment for Text-Video Retrievalcited by 103
  9. 2022TransVPR: Transformer-based place recognition with multi-level attention aggregationcited by 102
  10. 2023R2Former: Unified Retrieval and Reranking Transformer for Place Recognitioncited by 102
  11. 2022LaMAR: Benchmarking Localization and Mapping for Augmented Realitycited by 98
  12. 2022GSV-Cities: Toward Appropriate Supervised Visual Place Recognitioncited by 92
  13. 2022Beyond Cross-view Image Retrieval: Highly Accurate Vehicle Localization Using Satellite Imagecited by 88
  14. 2022Self-Supervised Learning for Videos: A Surveycited by 85
  15. 2023OrienterNet: Visual Localization in 2D Public Maps with Neural Matchingcited by 84
  16. 2023An Outlook into the Future of Egocentric Visioncited by 84
  17. 2022Deep Visual Geo-localization Benchmarkcited by 83
  18. 2021Where is your place, Visual Place Recognition?cited by 82
  19. 2024Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localizationcited by 81
  20. 2021PDC-Net+: Enhanced Probabilistic Dense Correspondence Networkcited by 80
  21. 2024How NeRFs and 3D Gaussian Splatting are Reshaping SLAM: a Surveycited by 72
  22. 2023Detector-Free Structure from Motioncited by 71
  23. 2021Deep Learning for Instance Retrieval: A Surveycited by 70
  24. 2021Transferable Visual Words: Exploiting the Semantics of Anatomical Patterns for Self-supervised Learningcited by 68
  25. 2021LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAMcited by 67
  26. 2023GlueStick: Robust Image Matching by Sticking Points and Lines Togethercited by 65
  27. 2021On the Limits of Pseudo Ground Truth in Visual Camera Re-localisationcited by 63
  28. 2024BoQ: A Place is Worth a Bag of Learnable Queriescited by 63
  29. 2022MeshLoc: Mesh-Based Visual Localizationcited by 62
  30. 2024CricaVPR: Cross-image Correlation-aware Representation Learning for Visual Place Recognitioncited by 61
  31. 2024CG-SLAM: Efficient Dense RGB-D SLAM in a Consistent Uncertainty-aware 3D Gaussian Fieldcited by 61
  32. 2024LoopSplat: Loop Closure by Registering 3D Gaussian Splatscited by 59
  33. 2025MegaLoc: One Retrieval to Place Them Allcited by 58
  34. 2021Semantics for Robotic Mapping, Perception and Interaction: A Surveycited by 56
  35. 2021SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognitioncited by 56
  36. 2021Fine-grained Semantics-aware Representation Enhancement for Self-supervised Monocular Depth Estimationcited by 55
  37. 2022A Transformer-Based Feature Segmentation and Region Alignment Method For UAV-View Geo-Localizationcited by 55
  38. 2021DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Featurescited by 54

50 most cited of 935 All 935 (JSON)