mlsanityData to 7 Oct 2026

On Label Granularity and Object Localization

arXiv 2207.10225 ↗First posted 20 Jul 2022cs.CV cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

Weakly supervised object localization (WSOL) aims to learn representations that encode object location using only image-level category labels. However, many objects can be labeled at different levels of granularity. Is it an animal, a bird, or a great horned owl? Which image-level labels should we use? In this paper we study the role of label granularity in WSOL. To facilitate this investigation we introduce iNatLoc500, a new large-scale fine-grained benchmark dataset for WSOL. Surprisingly, we find that choosing the right training label granularity provides a much larger performance boost than choosing the best WSOL algorithm. We also show that changing the label granularity can significantly improve data efficiency.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 46 of its 64 references that are in the map. First posted on arXiv 20 Jul 2022.

Builds on

46 of 64 references.

References · 46

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  5. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  6. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  7. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  8. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  9. 2013Fine-Grained Visual Classification of Aircraftcited by 1,701
  10. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,239
  11. 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 936
  12. 2018The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scalecited by 542
34 more
  1. 2020Are we done with ImageNet?cited by 329
  2. 2020FineGym: A Hierarchical Video Dataset for Fine-grained Action Understandingcited by 232
  3. 2017Hide-and-Seek: Forcing a Network to be Meticulous for Weakly-supervised Object and Action Localizationcited by 207
  4. 2015Weakly Supervised Deep Detection Networkscited by 183
  5. 2021Benchmarking Representation Learning for Natural World Image Collectionscited by 167
  6. 2018Adversarial Complementary Learning for Weakly Supervised Object Localizationcited by 166
  7. 2021Fine-Grained Image Analysis with Deep Learning: A Surveycited by 135
  8. 2019Attention-based Dropout Layer for Weakly Supervised Object Localizationcited by 125
  9. 2017Multiple Instance Detection Network with Online Instance Classifier Refinementcited by 117
  10. 2019Large-scale interactive object segmentation with human annotatorscited by 109
  11. 2019EdNet: A Large-Scale Hierarchical Dataset in Educationcited by 106
  12. 2020Evaluating Weakly Supervised Object Localization Methods Rightcited by 104
  13. 2021When Does Contrastive Visual Representation Learning Work?cited by 92
  14. 2021Weakly Supervised Object Localization and Detection: A Surveycited by 84
  15. 2017Extreme clicking for efficient object annotationcited by 80
  16. 2018Self-produced Guidance for Weakly-supervised Object Localizationcited by 72
  17. 2022Crafting Better Contrastive Views for Siamese Representation Learningcited by 67
  18. 2017Do Convolutional Neural Networks Learn Class Hierarchy?cited by 61
  19. 2016SqueezeDet: Unified, Small, Low Power Fully Convolutional Neural Networks for Real-Time Object Detection for Autonomous Drivingcited by 55
  20. 2021Towards Better Explanations of Class Activation Mappingcited by 47
  21. 2020Grafit: Learning fine-grained image representations with coarse labelscited by 43
  22. 2019The iMaterialist Fashion Attribute Datasetcited by 37
  23. 2020Concept Generalization in Visual Representation Learningcited by 35
  24. 2017Exploiting saliency for object segmentation from image level labelscited by 33
  25. 2019AnimalWeb: A Large-Scale Hierarchical Dataset of Annotated Animal Facescited by 28
  26. 2017Revisiting knowledge transfer for training object class detectorscited by 26
  27. 2020Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasetscited by 19
  28. 2021Normalization Matters in Weakly Supervised Object Localizationcited by 19
  29. 2020Strength from Weakness: Fast Learning Using Weak Supervisioncited by 18
  30. 2019Detecting 11K Classes: Large Scale Object Detection without Fine-Grained Bounding Boxescited by 17
  31. 2020In-sample Contrastive Learning and Consistent Attention for Weakly Supervised Object Localizationcited by 16
  32. 2021Keep CALM and Improve Visual Feature Attributioncited by 15
  33. 2020Weakly Supervised Representation Learning with Coarse Labelscited by 11
  34. 2020What leads to generalization of object proposals?cited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 4; 2025: 2; 2026: 2.

12 papers.

  1. 2023BioCLIP: A Vision Foundation Model for the Tree of Lifecited by 159
  2. 2023Bridging the Gap between Model Explanations in Partially Annotated Multi-label Classificationcited by 11
  3. 2024SHiNe: Semantic Hierarchy Nexus for Open-vocabulary Object Detectioncited by 11
  4. 2025Taxonomy-Aware Evaluation of Vision-Language Modelscited by 11
  5. 2023Image Labels Are All You Need for Coarse Seagrass Segmentationcited by 7
  6. 2024Labeled Data Selection for Category Discoverycited by 4
  7. 2023Open-World Weakly-Supervised Object Localizationcited by 2
  8. 2024Self-Masking Networks for Unsupervised Adaptationcited by 2
  9. 2025BackSplit: The Importance of Sub-dividing the Background in Biomedical Lesion Segmentationcited by 2
  10. 2024Graphical Perception of Saliency-based Model Explanationscited by 0
  11. 2026WildProp: Visual Estimation of Wildlife Body Proportions at Scalecited by 0
  12. 2026InterHier: Learning Interconnected Hierarchical Semantics for Open-Vocabulary Object Detectioncited by 0