Few-Shot Classification in Unseen Domains by Episodic Meta-Learning Across Visual Domains
This addresses the problem of few-shot classification in unseen domains for computer vision applications, representing an incremental advance over existing methods.
The paper tackles domain-generalized few-shot classification, where novel classes come from unseen target domains, by proposing a meta-learning framework that exploits multiple source domains to capture domain-invariant features. The result shows that learning from small, homogeneous source data performs preferably against large-scale data, with extensive experiments verifying effectiveness.
Few-shot classification aims to carry out classification given only few labeled examples for the categories of interest. Though several approaches have been proposed, most existing few-shot learning (FSL) models assume that base and novel classes are drawn from the same data domain. When it comes to recognizing novel-class data in an unseen domain, this becomes an even more challenging task of domain generalized few-shot classification. In this paper, we present a unique learning framework for domain-generalized few-shot classification, where base classes are from homogeneous multiple source domains, while novel classes to be recognized are from target domains which are not seen during training. By advancing meta-learning strategies, our learning framework exploits data across multiple source domains to capture domain-invariant features, with FSL ability introduced by metric-learning based mechanisms across support and query data. We conduct extensive experiments to verify the effectiveness of our proposed learning framework and show learning from small yet homogeneous source data is able to perform preferably against learning from large-scale one. Moreover, we provide insights into choices of backbone models for domain-generalized few-shot classification.