mlsanityData to 14 Oct 2026

Efficient Masked Attention Transformer for Few-Shot Classification and Segmentation

arXiv 2507.23642 ↗First posted 31 Jul 2025cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Few-shot classification and segmentation (FS-CS) focuses on jointly performing multi-label classification and multi-class segmentation using few annotated examples. Although the current state of the art (SOTA) achieves high accuracy in both tasks, it struggles with small objects. To overcome this, we propose the Efficient Masked Attention Transformer (EMAT), which improves classification and segmentation accuracy, especially for small objects. EMAT introduces three modifications: a novel memory-efficient masked attention mechanism, a learnable downscaling strategy, and parameter-efficiency enhancements. EMAT outperforms all FS-CS methods on the PASCAL-5i and COCO-20i datasets, using at least four times fewer trainable parameters. Moreover, as the current FS-CS evaluation setting discards available annotations, despite their costly collection, we introduce two novel evaluation settings that consider these annotations to better reflect practical scenarios.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 53 of its 66 references that are in the map. First posted on arXiv 31 Jul 2025.

Builds on

53 of 66 references.

References · 53

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  5. 2023Segment Anythingcited by 8,234
  6. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  7. 2023Visual Instruction Tuningcited by 7,869
  8. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  9. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  10. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
  11. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  12. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
38 more
  1. 2016Matching Networks for One Shot Learningcited by 2,255
  2. 2019PANet: Few-Shot Image Semantic Segmentation with Prototype Alignmentcited by 449
  3. 2020Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?cited by 437
  4. 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
  5. 2017One-Shot Learning for Semantic Segmentationcited by 337
  6. 2018How to train your MAMLcited by 309
  7. 2018Meta-Transfer Learning for Few-Shot Learningcited by 306
  8. 2019A Baseline for Few-Shot Image Classificationcited by 298
  9. 2020Prior Guided Feature Enrichment Network for Few-Shot Segmentationcited by 271
  10. 2021Hypercorrelation Squeeze for Few-Shot Segmentationcited by 193
  11. 2019Feature Weighting and Boosting for Few-Shot Segmentationcited by 189
  12. 2023Matcher: Segment Anything with One Shot Using All-Purpose Feature Matchingcited by 142
  13. 2019Infinite Mixture Prototypes for Few-Shot Learningcited by 99
  14. 2021Self-Guided and Cross-Guided Learning for Few-Shot Segmentationcited by 92
  15. 2023Hierarchical Dense Correlation Distillation for Few-Shot Segmentationcited by 84
  16. 2023Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinementcited by 84
  17. 2021Cross-domain Few-shot Learning with Task-specific Adapterscited by 80
  18. 2021Relational Embedding for Few-Shot Classificationcited by 74
  19. 2023A Closer Look at the Few-Shot Adaptation of Large Vision-Language Modelscited by 67
  20. 2020Generalized Few-shot Semantic Segmentationcited by 59
  21. 2023LLaFS: When Large Language Models Meet Few-Shot Segmentationcited by 48
  22. 2021Universal-Prototype Enhancing for Few-Shot Object Detectioncited by 43
  23. 2022Dynamic Prototype Convolution Network for Few-Shot Semantic Segmentationcited by 43
  24. 2019LaSO: Label-Set Operations networks for multi-label few-shot learningcited by 38
  25. 2024Bridge the Points: Graph-based Few-shot Segment Anything Semanticallycited by 35
  26. 2023Self-Calibrated Cross Attention Network for Few-Shot Segmentationcited by 32
  27. 2019Differentiable Meta-learning Model for Few-shot Semantic Segmentationcited by 31
  28. 2023Meta-Album: Multi-domain Meta-Dataset for Few-Shot Image Classificationcited by 30
  29. 2023Few-Shot Medical Image Segmentation via a Region-enhanced Prototypical Transformercited by 28
  30. 2022Integrative Few-Shot Learning for Classification and Segmentationcited by 25
  31. 2023MIANet: Aggregating Unbiased Instance and General Information for Few-Shot Semantic Segmentationcited by 25
  32. 2023Distilling Self-Supervised Vision Transformers for Weakly-Supervised Few-Shot Classification & Segmentationcited by 24
  33. 2024Adapt Before Comparison: A New Perspective on Cross-Domain Few-Shot Segmentationcited by 22
  34. 2021Meta-Learning for Multi-Label Few-Shot Classificationcited by 21
  35. 2024Rethinking Prior Information Generation with CLIP for Few-Shot Segmentationcited by 20
  36. 2022MSI: Maximize Support-Set Information for Few-Shot Segmentationcited by 18
  37. 2023Focus on Query: Adversarial Mining Transformer for Few-Shot Segmentationcited by 13
  38. 2020Deformable Gabor Feature Networks for Biomedical Image Classificationcited by 10

50 most cited of 53 All 53 (JSON)

Built on by

Not cited yet.