Skip to main content
Springer Nature Link
Log in

Improving Few-shot Learning by Spatially-aware Matching and CrossTransformer

  • Conference paper
  • First Online:
Computer Vision – ACCV 2022 (ACCV 2022)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 13845))

Included in the following conference series:

  • 430 Accesses

Abstract

Current few-shot learning models capture visual object relations in the so-called meta-learning setting under a fixed-resolution input. However, such models have a limited generalization ability under the scale and location mismatch between objects, as only few samples from target classes are provided. Therefore, the lack of a mechanism to match the scale and location between pairs of compared images leads to the performance degradation. The importance of image contents varies across coarse-to-fine scales depending on the object and its class label, e.g., generic objects and scenes rely on their global appearance while fine-grained objects rely more on their localized visual patterns. In this paper, we study the impact of scale and location mismatch in the few-shot learning scenario, and propose a novel Spatially-aware Matching (SM) scheme to effectively perform matching across multiple scales and locations, and learn image relations by giving the highest weights to the best matching pairs. The SM is trained to activate the most related locations and scales between support and query data. We apply and evaluate SM on various few-shot learning models and backbones for comprehensive evaluations. Furthermore, we leverage an auxiliary self-supervisory discriminator to train/predict the spatial- and scale-level index of feature vectors we use. Finally, we develop a novel transformer-based pipeline to exploit self- and cross-attention in a spatially-aware matching process. Our proposed design is orthogonal to the choice of backbone and/or comparator.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

References

  1. Vinyals, O., Blundell, C., Lillicrap, T., Wierstra, D., et al.: Matching networks for one shot learning. In: NIPS, pp. 3630–3638 (2016)

    Google Scholar 

  2. Snell, J., Swersky, K., Zemel, R.: Prototypical networks for few-shot learning. In: NeurIPS, pp. 4077–4087 (2017)

    Google Scholar 

  3. Sung, F., Yang, Y., Zhang, L., Xiang, T., Torr, P.H., Hospedales, T.M.: Learning to compare: Relation network for few-shot learning. arXiv preprint arXiv:1711.06025 (2017)

  4. Zhang, H., Koniusz, P.: Power normalizing second-order similarity network for few-shot learning. In: WACV, pp. 1185–1193. IEEE (2019)

    Google Scholar 

  5. Porikli, F., Tuzel, O.: Covariance tracker. In: CVPR (2006)

    Google Scholar 

  6. Guo, K., Ishwar, P., Konrad, J.: Action recognition from video using feature covariance matrices. Trans. Imgage Process. 22, 2479–2494 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  7. Carreira, J., Caseiro, R., Batista, J., Sminchisescu, C.: Semantic segmentation with second-order pooling. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012. LNCS, vol. 7578, pp. 430–443. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-33786-4_32

    Chapter  Google Scholar 

  8. Koniusz, P., Yan, F., Gosselin, P.H., Mikolajczyk, K.: Higher-order occurrence pooling for bags-of-words: visual concept detection. IEEE Trans. Pattern Anal. Mach. Intell. 39, 313–326 (2017)

    Article  Google Scholar 

  9. Koniusz, P., Zhang, H., Porikli, F.: A deeper look at power normalizations. In: CVPR, pp. 5774–5783 (2018)

    Google Scholar 

  10. Wertheimer, D., Hariharan, B.: Few-shot learning with localization in realistic settings. In: CVPR, pp. 6558–6567 (2019)

    Google Scholar 

  11. Lazebnik, S., Schmid, C., Ponce, J.: Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories. CVPR 2, 2169–2178 (2006)

    Google Scholar 

  12. Yang, J., Yu, K., Gong, Y., Huang, T.: Linear spatial pyramid matching using sparse coding for image classification. In: CVPR, pp. 1794–1801 (2009)

    Google Scholar 

  13. Miller, E.G., Matsakis, N.E., Viola, P.A.: Learning from one example through shared densities on transforms. In: CVPR, vol. 1, pp. 464–471 (2000)

    Google Scholar 

  14. Li, F.F., VanRullen, R., Koch, C., Perona, P.: Rapid natural scene categorization in the near absence of attention. Proc. Natl. Acad. Sci. 99, 9596–9601 (2002)

    Article  Google Scholar 

  15. Fink, M.: Object classification from a single example utilizing class relevance metrics. In: NIPS, pp. 449–456 (2005)

    Google Scholar 

  16. Bart, E., Ullman, S.: Cross-generalization: Learning novel classes from a single example by feature replacement. In: CVPR, pp. 672–679 (2005)

    Google Scholar 

  17. Fei-Fei, L., Fergus, R., Perona, P.: One-shot learning of object categories. IEEE Trans. Pattern Anal. Mach. Intell. 28, 594–611 (2006)

    Article  Google Scholar 

  18. Lake, B.M., Salakhutdinov, R., Gross, J., Tenenbaum, J.B.: One shot learning of simple visual concepts. CogSci (2011)

    Google Scholar 

  19. Koch, G., Zemel, R., Salakhutdinov, R.: Siamese neural networks for one-shot image recognition. In: ICML Deep Learning Workshop, vol. 2 (2015)

    Google Scholar 

  20. Finn, C., Abbeel, P., Levine, S.: Model-agnostic meta-learning for fast adaptation of deep networks. In: ICML, pp. 1126–1135 (2017)

    Google Scholar 

  21. Garcia, V., Bruna, J.: Few-shot learning with graph neural networks. arXiv preprint arXiv:1711.04043 (2017)

  22. Rusu, A.A., et al.: Meta-learning with latent embedding optimization. arXiv preprint arXiv:1807.05960 (2018)

  23. Gidaris, S., Komodakis, N.: Dynamic few-shot visual learning without forgetting. In: CVPR, pp. 4367–4375 (2018)

    Google Scholar 

  24. Zhang, H., Zhang, J., Koniusz, P.: Few-shot learning via saliency-guided hallucination of samples. In: CVPR, pp. 2770–2779 (2019)

    Google Scholar 

  25. Kim, J., Kim, T., Kim, S., Yoo, C.D.: Edge-labeling graph neural network for few-shot learning. In: CVPR (2019)

    Google Scholar 

  26. Gidaris, S., Komodakis, N.: Generating classification weights with GNN denoising autoencoders for few-shot learning. In: CVPR (2019)

    Google Scholar 

  27. Hou, R., Chang, H., Ma, B., Shan, S., Chen, X.: Cross attention network for few-shot classification. In: NeurIPS, vol. 32 (2019)

    Google Scholar 

  28. Wu, Z., Li, Y., Guo, L., Jia, K.: Parn: Position-aware relation networks for few-shot learning. In: ICCV, pp. 6659–6667 (2019)

    Google Scholar 

  29. Hao, F., He, F., Cheng, J., Wang, L., Cao, J., Tao, D.: Collect and select: semantic alignment metric learning for few-shot learning. In: CVPR, pp. 8460–8469 (2019)

    Google Scholar 

  30. Li, W., Wang, L., Xu, J., Huo, J., Gao, Y., Luo, J.: Revisiting local descriptor based image-to-class measure for few-shot learning. In: CVPR, pp. 7260–7268 (2019)

    Google Scholar 

  31. Zhang, H., Li, H., Koniusz, P.: Multi-level second-order few-shot learning. IEEE Trans. Multimed. 99, 1–16 (2022)

    Google Scholar 

  32. Ni, G., Zhang, H., Zhao, J., Xu, L., Yang, W., Lan, L.: ANF: attention-based noise filtering strategy for unsupervised few-shot classification. In: Pham, D.N., Theeramunkong, T., Governatori, G., Liu, F. (eds.) PRICAI 2021. LNCS (LNAI), vol. 13033, pp. 109–123. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-89370-5_9

    Chapter  Google Scholar 

  33. Simon, C., Koniusz, P., Nock, R., Harandi, M.: On modulating the gradient for meta-learning. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12353, pp. 556–572. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58598-3_33

    Chapter  Google Scholar 

  34. Sun, K., Koniusz, P., Wang, Z.: Fisher-Bures adversary graph convolutional networks. UAI 115, 465–475 (2019)

    Google Scholar 

  35. Zhu, H., Koniusz, P.: Simple spectral graph convolution. In: ICLR (2021)

    Google Scholar 

  36. Zhu, H., Sun, K., Koniusz, P.: Contrastive Laplacian eigenmaps. In: NeurIPS, pp. 5682–5695 (2021)

    Google Scholar 

  37. Zhang, Y., Zhu, H., Meng, Z., Koniusz, P., King, I.: Graph-adaptive rectified linear unit for graph neural networks. In: The Web Conference (WWW), pp. 1331–1339. ACM (2022)

    Google Scholar 

  38. Zhang, Y., Zhu, H., Song, Z., Koniusz, P., King, I.: COSTA: covariance-preserving feature augmentation for graph contrastive learning. In: KDD, pp. 2524–2534. ACM (2022)

    Google Scholar 

  39. Wang, L., Liu, J., Koniusz, P.: 3d skeleton-based few-shot action recognition with JEANIE is not so naïve. arXiv preprint arXiv: 2112.12668 (2021)

  40. Wang, L., Koniusz, P.: Temporal-viewpoint transportation plan for skeletal few-shot action recognition. In: ACCV (2022)

    Google Scholar 

  41. Wang, L., Koniusz, P.: Uncertainty-DTW for time series and sequences. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds.) Computer Vision–ECCV 2022. ECCV 2022. LNCS, vol. 13681, pp. 176–195. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-19803-8_11

  42. Kang, D., Kwon, H., Min, J., Cho, M.: Relational embedding for few-shot classification. In: ICCV, pp. 8822–8833 (2021)

    Google Scholar 

  43. Zhu, H., Koniusz, P.: EASE: unsupervised discriminant subspace learning for transductive few-shot learning. In: CVPR (2022)

    Google Scholar 

  44. Lu, C., Koniusz, P.: Few-shot keypoint detection with uncertainty learning for unseen species. In: CVPR (2022)

    Google Scholar 

  45. Tuzel, O., Porikli, F., Meer, P.: Region covariance: a fast descriptor for detection and classification. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006. LNCS, vol. 3952, pp. 589–600. Springer, Heidelberg (2006). https://doi.org/10.1007/11744047_45

    Chapter  Google Scholar 

  46. Koniusz, P., Cherian, A., Porikli, F.: Tensor representations via kernel linearization for action recognition from 3d skeletons. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9908, pp. 37–53. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46493-0_3

    Chapter  Google Scholar 

  47. Koniusz, P., Wang, L., Cherian, A.: Tensor representations for action recognition. IEEE Trans. Pattern Anal. Mach. Intell. 44, 648–665 (2022)

    Article  Google Scholar 

  48. Koniusz, P., Tas, Y., Porikli, F.: Domain adaptation by mixture of alignments of second-or higher-order scatter tensors. In: CVPR, vol. 2 (2017)

    Google Scholar 

  49. Tas, Y., Koniusz, P.: CNN-based action recognition and supervised domain adaptation on 3d body skeletons via kernel feature maps. In: BMVC, p. 158. BMVA Press (2018)

    Google Scholar 

  50. Zhang, H., Koniusz, P., Jian, S., Li, H., Torr, P.H.S.: Rethinking class relations: absolute-relative supervised and unsupervised few-shot learning. In: CVPR, pp. 9432–9441 (2021)

    Google Scholar 

  51. Koniusz, P., Zhang, H.: Power normalizations in fine-grained image, few-shot image and graph classification. IEEE Trans. Pattern Anal. Mach. Intell. 44, 591–609 (2022)

    Article  Google Scholar 

  52. Zhang, S., Luo, D., Wang, L., Koniusz, P.: Few-shot object detection by second-order pooling. In: Ishikawa, H., Liu, C.-L., Pajdla, T., Shi, J. (eds.) ACCV 2020. LNCS, vol. 12625, pp. 369–387. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-69538-5_23

    Chapter  Google Scholar 

  53. Yu, X., Zhuang, Z., Koniusz, P., Li, H.: 6DoF object pose estimation via differentiable proxy voting regularizer. In: BMVC, BMVA Press (2020)

    Google Scholar 

  54. Zhang, S., Wang, L., Murray, N., Koniusz, P.: Kernelized few-shot object detection with efficient integral aggregation. In: CVPR, pp. 19207–19216 (2022)

    Google Scholar 

  55. Zhang, S., Murray, N., Wang, L., Koniusz, P.: Time-rEversed DiffusioN tEnsor Transformer: a new TENET of few-shot object detection. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds.) Computer Vision–ECCV 2022. ECCV 2022. LNCS, vol. 13680, pp. 310–328. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-20044-1_18

  56. Simon, C., Koniusz, P., Harandi, M.: On learning the geodesic path for incremental learning. In: CVPR, pp. 1591–1600 (2021)

    Google Scholar 

  57. Doersch, C., Gupta, A., Zisserman, A.: Crosstransformers: spatially-aware few-shot transfer. arXiv preprint arXiv:2007.11498 (2020)

  58. Antoniou, A., Edwards, H., Storkey, A.: How to train your maml. arXiv preprint arXiv:1810.09502 (2018)

  59. Oreshkin, B., Lopez, P.R., Lacoste, A.: TADAM: task dependent adaptive metric for improved few-shot learning. In: Advances in Neural Information Processing Systems, pp. 721–731 (2018)

    Google Scholar 

  60. Lee, K., Maji, S., Ravichandran, A., Soatto, S.: Meta-learning with differentiable convex optimization. In: CVPR, pp. 10657–10665 (2019)

    Google Scholar 

  61. Zhang, C., Cai, Y., Lin, G., Shen, C.: DeepEMD: few-shot image classification with differentiable earth mover’s distance and structured classifiers. In: CVPR, pp. 12203–12213 (2020)

    Google Scholar 

  62. Triantafillou, E., et al.: Meta-dataset: A dataset of datasets for learning to learn from few examples. arXiv preprint arXiv:1903.03096 (2019)

  63. Ren, M., et al.: Meta-learning for semi-supervised few-shot classification. In: ICLR (2018)

    Google Scholar 

  64. Koniusz, P., Tas, Y., Zhang, H., Harandi, M., Porikli, F., Zhang, R.: Museum exhibit identification challenge for the supervised domain adaptation and beyond. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11220, pp. 815–833. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01270-0_48

    Chapter  Google Scholar 

  65. Nilsback, M.E., Zisserman, A.: Automated flower classification over a large number of classes. In: Proceedings of the Indian Conference on Computer Vision, Graphics and Image Processing (2008)

    Google Scholar 

  66. Wah, C., Branson, S., Welinder, P., Perona, P., Belongie, S.: The Caltech-UCSD Birds-200-2011 Dataset. Technical report CNS-TR-2011-001, California Institute of Technology (2011)

    Google Scholar 

  67. Bossard, L., Guillaumin, M., Van Gool, L.: Food-101 – mining discriminative components with random forests. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8694, pp. 446–461. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10599-4_29

    Chapter  Google Scholar 

Download references

Acknowledgements

This work is supported by National Natural Science Fundation of China (No. 62106282), and Young Elite Scientists Sponsorship Program by CAST (No. 2021-JCJQ-QT-038). Code: https://github.com/HongguangZhang/smfsl-master.

Author information

Authors and Affiliations

  1. Systems Engineering Institute, AMS, Shanghai, China

    Hongguang Zhang

  2. Data61/CSIRO, Sydney, Australia

    Piotr Koniusz

  3. Australian National University, Canberra, Australia

    Piotr Koniusz

  4. Oxford University, Oxford, UK

    Philip H. S. Torr

Authors
  1. Hongguang Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Philip H. S. Torr
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Piotr Koniusz
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Hongguang Zhang .

Editor information

Editors and Affiliations

  1. University of Wollongong, Wollongong, NSW, Australia

    Lei Wang

  2. University of Bonn, Bonn, Germany

    Juergen Gall

  3. University of Adelaide, Adelaide, SA, Australia

    Tat-Jun Chin

  4. National Institute of Informatics, Tokyo, Japan

    Imari Sato

  5. Johns Hopkins University, Baltimore, MD, USA

    Rama Chellappa

Rights and permissions

Reprints and permissions

Copyright information

© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Zhang, H., Torr, P.H.S., Koniusz, P. (2023). Improving Few-shot Learning by Spatially-aware Matching and CrossTransformer. In: Wang, L., Gall, J., Chin, TJ., Sato, I., Chellappa, R. (eds) Computer Vision – ACCV 2022. ACCV 2022. Lecture Notes in Computer Science, vol 13845. Springer, Cham. https://doi.org/10.1007/978-3-031-26348-4_1

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-26348-4_1

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-26347-7

  • Online ISBN: 978-3-031-26348-4

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics