mlsanityData to 14 Oct 2026

Learning Primitive-aware Discriminative Representations for Few-shot Learning

arXiv 2208.09717 ↗First posted 20 Aug 2022cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Few-shot learning (FSL) aims to learn a classifier that can be easily adapted to recognize novel classes with only a few labeled examples. Some recent work about FSL has yielded promising classification performance, where the image-level feature is used to calculate the similarity among samples for classification. However, the image-level feature ignores abundant fine-grained and structural in-formation of objects that may be transferable and consistent between seen and unseen classes. How can humans easily identify novel classes with several sam-ples? Some study from cognitive science argues that humans can recognize novel categories through primitives. Although base and novel categories are non-overlapping, they can share some primitives in common. Inspired by above re-search, we propose a Primitive Mining and Reasoning Network (PMRN) to learn primitive-aware representations based on metric-based FSL model. Concretely, we first add Self-supervision Jigsaw task (SSJ) for feature extractor parallelly, guiding the model to encode visual pattern corresponding to object parts into fea-ture channels. To further mine discriminative representations, an Adaptive Chan-nel Grouping (ACG) method is applied to cluster and weight spatially and se-mantically related visual patterns to generate a group of visual primitives. To fur-ther enhance the discriminability and transferability of primitives, we propose a visual primitive Correlation Reasoning Network (CRN) based on graph convolu-tional network to learn abundant structural information and internal correlation among primitives. Finally, a primitive-level metric is conducted for classification in a meta-task based on episodic training strategy. Extensive experiments show that our method achieves state-of-the-art results on six standard benchmarks.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 44 of its 67 references that are in the map. First posted on arXiv 20 Aug 2022.

Builds on

44 of 67 references.

References · 44

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  5. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  6. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  7. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  8. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  9. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
  10. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  11. 2015Learning Deep Features for Discriminative Localizationcited by 2,436
  12. 2016Matching Networks for One Shot Learningcited by 2,255
32 more
  1. 2013Fine-Grained Visual Classification of Aircraftcited by 1,726
  2. 2016Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzlescited by 1,382
  3. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,143
  4. 2017Network Dissection: Quantifying Interpretability of Deep Visual Representationscited by 708
  5. 2018Dynamic Few-Shot Visual Learning without Forgettingcited by 385
  6. 2017Few-Shot Learning with Graph Neural Networkscited by 324
  7. 2018How to train your MAMLcited by 309
  8. 2018Meta-Transfer Learning for Few-Shot Learningcited by 306
  9. 2018Few-Shot Learning via Embedding Adaptation with Set-to-Set Functionscited by 294
  10. 2019Cross Attention Network for Few-shot Classificationcited by 219
  11. 2018Self-Supervised Video Representation Learning with Space-Time Cubic Puzzlescited by 149
  12. 2019Boosting Few-Shot Visual Learning with Self-Supervisioncited by 140
  13. 2019Charting the Right Manifold: Manifold Mixup for Few-shot Learningcited by 136
  14. 2018Net2Vec: Quantifying and Explaining how Concepts are Encoded by Filters in Deep Neural Networkscited by 129
  15. 2019Revisiting Local Descriptor based Image-to-Class Measure for Few-shot Learningcited by 121
  16. 2020Few-Shot Classification with Feature Map Reconstruction Networkscited by 116
  17. 2019Finding Task-Relevant Features for Few-Shot Learning by Category Traversalcited by 101
  18. 2019When Does Self-supervision Improve Few-shot Learning?cited by 78
  19. 2018Memory Matching Networks for One-Shot Image Recognitioncited by 65
  20. 2019Associative Alignment for Few-shot Image Classificationcited by 58
  21. 2019Dense Classification and Implanting for Few-Shot Learningcited by 56
  22. 2017Few-Shot Learning Through an Information Retrieval Lenscited by 51
  23. 2018Learning Compositional Representations for Few-Shot Recognitioncited by 50
  24. 2020DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learningcited by 49
  25. 2019Low-Rank Pairwise Alignment Bilinear Network For Few-Shot Fine-Grained Image Classificationcited by 36
  26. 2022Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-shot Learningcited by 27
  27. 2018Deep Meta-Learning: Learning to Learn in the Concept Spacecited by 22
  28. 2021Learning to Affiliate: Mutual Centralized Learning for Few-shot Classificationcited by 18
  29. 2019Diversity Transfer Network for Few-Shot Learningcited by 15
  30. 2020Multi-scale Adaptive Task Attention Network for Few-Shot Learningcited by 10
  31. 2021Multi-level Metric Learning for Few-shot Image Recognitioncited by 8
  32. 2019Weakly-supervised Compositional FeatureAggregation for Few-shot Recognitioncited by 3

Built on by

Not cited yet.