Mutual Correlation Network for few-shot learning.

Journal: Neural networks : the official journal of the International Neural Network Society
Published Date:

Abstract

Most metric-based Few-Shot Learning (FSL) methods focus on learning good embeddings of images. However, these methods either lack the ability to explore the cross-correlation (i.e., correlated information) between image pairs or explore limited consensus among the correlation map constrained by the limited receptive field of CNN. We propose a Mutual Correlation Network (MCNet) to explore global consensus among the correlation map by using the self-attention mechanism which has a global receptive field. Our MCNet contains two core modules: (1) a multi-level embedding module that generates multi-level embeddings for an image pair which capture hierarchical semantics, and (2) a mutual correlation module that refines correlation map of two embeddings and generates more robust relational embeddings. Extensive experiments show that our MCNet achieves competitive results on four widely-used few-shot classification benchmarks miniImageNet, tieredImageNet, CUB-200-2011, and CIFAR-FS. Code is available at https://github.com/DRGreat/MCNet.

Authors

  • Derong Chen
    School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China.
  • Feiyu Chen
    Department of Electrical Engineering, Stanford University, Stanford, California, USA.
  • Deqiang Ouyang
    Center for Future Media, School of Computer Science & Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China. Electronic address: ouyangdeqiang@std.uestc.edu.cn.
  • Jie Shao
    Center for Future Media, School of Computer Science & Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China; Sichuan Artificial Intelligence Research Institute, Yibin 644000, China. Electronic address: shaojie@uestc.edu.cn.