Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Journal of Image Processing Theory and Applications
  4. Vol 7, Issue 1, 2024

Few-shot Image Classification Model Based on Improved Prototype

Download as PDF

DOI: 10.23977/jipta.2024.070115 | Downloads: 18 | Views: 791

Author(s)

Ya Gao 1, Chunhui Liang 1, Jianye An 1

Affiliation(s)

1 School of Science, Tianjin University of Commerce, Tianjin, 300134, China

Corresponding Author

Ya Gao

ABSTRACT

Aiming at the limitations of important feature extraction in few-shot image classification and the problem of poor representation of category prototypes in the prototype network, this paper proposes a few-shot image classification model based on improved prototype. The model adds a global grouping multi-attention mechanism to the original backbone network in the feature extraction module, which enhances the extraction of important features. For the prototype improvement module, the pseudo-labeled samples in the query set with cosine similarity scores higher than a threshold are utilized to expand the support set. Finally, applying our model to the typical few-shot image datasets miniImagenet and tieredImagenet, the experiments show that compared with other meta-learning models, the model built in this paper achieves better classification performance.

KEYWORDS

few-shot image classification, meta-learning, muti-head attentional mechanism, pseudo-labeled sample expansion

CITE THIS PAPER

Ya Gao, Chunhui Liang, Jianye An, Few-shot Image Classification Model Based on Improved Prototype. Journal of Image Processing Theory and Applications (2024) Vol. 7: 125-133. DOI: http://dx.doi.org/10.23977/jipta.2024.070115.

REFERENCES

[1] Zhou B.J, Chen C.Y. A review of research on few-shot image classification based on deep meta-learning [J]. Computer Engineering and Applications, 2024, 60(08):1-15.
[2] Song Y, Wang T, Cai P, et al. A comprehensive survey of few-shot learning: Evolution, applications, challenges, and opportunities [J]. ACM Computing Surveys, 2023, 55(13s): 1-40.
[3] Hu Y , Pateux S , Gripon V .Squeezing Backbone Feature Distributions to the Max for Efficient Few-Shot Learning[J]. algorithms, 2022, 15(5).
[4] Liu Y, Zhang H, Yang Y. Few-Shot Image Classification Based on Asymmetric Convolution and Attention Mechanism[C/OL]//2022 4th International Conference on Natural Language Processing (ICNLP), Xi’an, China. 2022:217-222
[5] Mikolov T, Martin Karafiát, Burget L ,et al. Recurrent neural network based language model[C]//Interspeech, Conference of the International Speech Communication Association, Makuhari, Chiba, Japan, September.DBLP, 2015.
[6] Sergey B, Adam S, Matthew B,et al. Meta-Learning with Memory-Augmented Neural Networks[J].Journal of Machine Learning Research, 2016.
[7] Cai Q, Pan Y, Yao T,et al. Memory Matching Networks for One-Shot Image Recognition[J].IEEE, 2018:4080-4088.
[8] Finn C, Abbeel P, Levine S. Model-agnostic meta-learning for fast adaptation of deep networks[C]// International conference on machine learning. PMLR, 2017: 1126-1135.
[9] Snell J, Swersky K, Zemel R S. Prototypical networks for few-shot learning[J]. Advances in neural information processing systems, 2017, 30.
[10] Sung F, Yang Y, Zhang L, et al. Learning to compare: relation network for few-shot learning[C]//Proceedings of the 31st IEEE Conference on Computer Vision and Pattern Recognition(CVPR’18), 2018: 1199-1208.
[11] Liu Y, Lee J, Park M, et al. Transductive propagation network for few-shot learning [J]. arXiv preprint arXiv:1805.10002, 2018.
[12] Rusu, Andrei A A, Rao D, Sygnowski J, et al. Meta-learning with latent embedding optimization[J]. arXiv preprint arXiv:1807.05960, 2018. 
[13] Lee K, Maji S, Ravichandran A, et al. Meta-learning with differentiable convex optimization[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2019:10649-10657.
[14] Zhong X, Gu C, Huang W,et al. Complementing Representation Deficiency in Few-shot Image Classification: A Meta-Learning Approach[C]//International Conference on Pattern Recognition. IEEE, 2021: 2677-2684.
[15] Zhou F, Zhang L, Wei W. Meta-generating deep attentive metric for few-shot classification[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2022: 6863-6873.
[16] Chen C , Li K , Wei W ,et al. Hierarchical Graph Neural Networks for Few-Shot Learning[J].IEEE Transactions on Circuits and Systems for Video Technology, 2021:240-252.
[17] Liu J, Song L, Qin Y. Prototype Rectification for Few-Shot Learning[C]//European Conference on Computer Vision.Springer, Cham, 2020:741-756.
[18] Ahmed M, Seraj R, Islam S M S .The k-means Algorithm: A Comprehensive Survey and Performance Evaluation[J]. Electronics, 2020, 9(8):1295.
[19] Russakovsky O, Deng J, Su H, et al. Imagenet large scale visual recognition challenge[J]. International journal of computer vision, 2015, 115: 211-252.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 2455
Visits: 172158

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.