mlsanityData to 7 Oct 2026

Matching Feature Sets for Few-Shot Image Classification

arXiv 2204.00949 ↗First posted 2 Apr 2022cs.CVcited by 25

Not on the map. Why?

Abstract (arXiv)

In image classification, it is common practice to train deep networks to extract a single feature vector per input image. Few-shot classification methods also mostly follow this trend. In this work, we depart from this established direction and instead propose to extract sets of feature vectors for each image. We argue that a set-based representation intrinsically builds a richer representation of images from the base classes, which can subsequently better transfer to the few-shot classes. To do so, we propose to adapt existing feature extractors to instead produce sets of feature vectors from images. Our approach, dubbed SetFeat, embeds shallow self-attention mechanisms inside existing encoder architectures. The attention modules are lightweight, and as such our method results in encoders that have approximately the same number of parameters as their original versions. During training and inference, a set-to-set matching metric is used to perform image classification. The effectiveness of our proposed architecture and metrics is demonstrated via thorough experiments on standard few-shot datasets -- namely miniImageNet, tieredImageNet, and CUB -- in both the 1- and 5-shot scenarios. In all cases but one, our method outperforms the state-of-the-art.

Key facts. Cited by 25 papers in the map (data through 7 Oct 2026). Builds on 58 of its 77 references that are in the map. First posted on arXiv 2 Apr 2022.

Builds on

58 of 77 references.

References · 58

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  5. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  6. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  7. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  8. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  9. 2016Matching Networks for One Shot Learningcited by 2,246
  10. 2017Deep Setscited by 1,394
  11. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
  12. 2021CvT: Introducing Convolutions to Vision Transformerscited by 946
38 more
  1. 2019A Closer Look at Few-shot Classificationcited by 679
  2. 2021LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inferencecited by 475
  3. 2019Meta-Learning with Differentiable Convex Optimizationcited by 461
  4. 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
  5. 2018TADAM: Task dependent adaptive metric for improved few-shot learningcited by 408
  6. 2018Meta-learning with differentiable closed-form solverscited by 395
  7. 2018Dynamic Few-Shot Visual Learning without Forgettingcited by 385
  8. 2017Few-Shot Learning with Graph Neural Networkscited by 324
  9. 2018How to train your MAMLcited by 306
  10. 2018Meta-Transfer Learning for Few-Shot Learningcited by 306
  11. 2019A Baseline for Few-Shot Image Classificationcited by 298
  12. 2018Few-Shot Learning via Embedding Adaptation with Set-to-Set Functionscited by 293
  13. 2020Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effectcited by 256
  14. 2018Probabilistic Model-Agnostic Meta-Learningcited by 254
  15. 2019Cross Attention Network for Few-shot Classificationcited by 218
  16. 2018Bayesian Model-Agnostic Meta-Learningcited by 194
  17. 2020Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformationcited by 183
  18. 2018Low-Shot Learning from Imaginary Datacited by 181
  19. 2020Meta-Baseline: Exploring Simple Meta-Learning for Few-Shot Learningcited by 179
  20. 2020CrossTransformers: spatially-aware few-shot transfercited by 174
  21. 2019Boosting Few-Shot Visual Learning with Self-Supervisioncited by 140
  22. 2020Negative Margin Matters: Understanding Margin in Few-shot Classificationcited by 129
  23. 2019Revisiting Local Descriptor based Image-to-Class Measure for Few-shot Learningcited by 121
  24. 2019Infinite Mixture Prototypes for Few-Shot Learningcited by 99
  25. 2019TapNet: Neural Network Augmented with Task-Adaptive Projection for Few-Shot Learningcited by 90
  26. 2018Delta-encoder: an effective sample synthesis method for few-shot object recognitioncited by 89
  27. 2019When Does Self-supervision Improve Few-shot Learning?cited by 78
  28. 2019Diversity with Cooperation: Ensemble Methods for Few-Shot Classificationcited by 70
  29. 2019Generating Classification Weights with GNN Denoising Autoencoders for Few-Shot Learningcited by 69
  30. 2018Memory Matching Networks for One-Shot Image Recognitioncited by 65
  31. 2019Associative Alignment for Few-shot Image Classificationcited by 58
  32. 2019Dense Classification and Implanting for Few-Shot Learningcited by 56
  33. 2019Image Deformation Meta-Networks for One-Shot Learningcited by 56
  34. 2021Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot Learningcited by 56
  35. 2020Prototype Completion with Primitive Knowledge for Few-Shot Learningcited by 50
  36. 2019Few-Shot Learning via Saliency-guided Hallucination of Samplescited by 49
  37. 2020Transductive Information Maximization For Few-Shot Learningcited by 44
  38. 2021Learning Dynamic Alignment via Meta-filter for Few-shot Learningcited by 41

50 most cited of 58 All 58 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 14; 2024: 4; 2025: 3; 2026: 1.

25 papers.

Sort:

By area:

  1. 2023Semantic Prompt for Few-Shot Image Recognitioncited by 29
  2. 2023Exploring Incompatible Knowledge Transfer in Few-shot Image Generationcited by 19
  3. 2024Compositional Few-Shot Class-Incremental Learningcited by 18
  4. 2023ProtoDiff: Learning to Learn Prototypical Networks by Task-Guided Diffusioncited by 15
  5. 2023DETA: Denoised Task Adaptation for Few-Shot Learningcited by 13
  6. 2023Reference Twice: A Simple and Unified Baseline for Few-Shot Instance Segmentationcited by 12
  7. 2022An Embarrassingly Simple Approach to Semi-Supervised Few-Shot Learningcited by 10
  8. 2023Prompt Tuning based Adapter for Vision-Language Model Adaptioncited by 10
  9. 2023Mitigating the Effect of Incidental Correlations on Part-based Learningcited by 6
  10. 2023The Art of Camouflage: Few-Shot Learning for Animal Detection and Segmentationcited by 5
  11. 2023Instance-based Max-margin for Practical Few-shot Recognitioncited by 5
  12. 2022Enhancing Few-shot Image Classification with Cosine Transformercited by 4
13 more
  1. 2023Meta Omnium: A Benchmark for General-Purpose Learning-to-Learncited by 4
  2. 2023Few-shot Classification via Ensemble Learning with Multi-Order Statisticscited by 3
  3. 2025Contrastive Visual Data Augmentationcited by 3
  4. 2022Robust Meta-Representation Learning via Global Label Inference and Classificationcited by 2
  5. 2023Dual Adversarial Alignment for Realistic Support-Query Shift Few-shot Learningcited by 1
  6. 2024Dual-View Data Hallucination with Semantic Relation Guidance for Few-Shot Image Recognitioncited by 1
  7. 2024Task Attribute Distance for Few-Shot Learning: Theoretical Analysis and Applicationscited by 1
  8. 2023Recent Advances of Local Mechanisms in Computer Vision: A Survey and Outlook of Recent Workcited by 0
  9. 2023Few-shot Image Classification based on Gradual Machine Learningcited by 0
  10. 2024Like Humans to Few-Shot Learning through Knowledge Permeation of Vision and Textcited by 0
  11. 2025Brain Inspired Adaptive Memory Dual-Net for Few-Shot Image Classificationcited by 0
  12. 2025MetaChest: Generalized few-shot learning of pathologies from chest X-rayscited by 0
  13. 2026PMCE: Probabilistic Multi-Granularity Semantics with Caption-Guided Enhancement for Few-Shot Learningcited by 0