Binocular Mutual Learning for Improving Few-shot Classification

TL;DR

This paper introduces Binocular Mutual Learning (BML), a unified framework that combines global and local views for few-shot classification, leveraging mutual interaction to enhance learning and improve accuracy across benchmarks.

Contribution

The paper proposes a novel BML framework that integrates global and local class views through intra-view and cross-view interactions for better few-shot learning.

Findings

BML outperforms existing methods on multiple benchmarks.

Cross-view mutual interaction improves classification accuracy.

Binocular embeddings effectively support decision-making during meta-test.

Abstract

Most of the few-shot learning methods learn to transfer knowledge from datasets with abundant labeled data (i.e., the base set). From the perspective of class space on base set, existing methods either focus on utilizing all classes under a global view by normal pretraining, or pay more attention to adopt an episodic manner to train meta-tasks within few classes in a local view. However, the interaction of the two views is rarely explored. As the two views capture complementary information, we naturally think of the compatibility of them for achieving further performance gains. Inspired by the mutual learning paradigm and binocular parallax, we propose a unified framework, namely Binocular Mutual Learning (BML), which achieves the compatibility of the global view and the local view through both intra-view and cross-view modeling. Concretely, the global view learns in the whole class space to capture rich inter-class relationships. Meanwhile, the local view learns in the local class space within each episode, focusing on matching positive pairs correctly. In addition, cross-view mutual interaction further promotes the collaborative learning and the implicit exploration of useful knowledge from each other. During meta-test, binocular embeddings are aggregated together to support decision-making, which greatly improve the accuracy of classification. Extensive experiments conducted on multiple benchmarks including cross-domain validation confirm the effectiveness of our method.