Revisiting Mid-Level Patterns for Distant-Domain Few-Shot Recognition

Existing few-shot learning (FSL) methods usually assume known classes and novel classes are from the same domain (in-domain setting). However in practice, it may be infeasible to collect sufficient training samples for some special domains to construct known classes. To solve this problem, cross-domain FSL (CDFSL) is proposed very recently to transfer knowledge from general-domain known classes to special-domain novel classes. Existing CDFSL works mostly focus on transferring between close domains, while rarely consider transferring between distant domains, which is even more challenging. In this paper, we study distant-domain FSL, a challenging subset of CDFSL, by revisiting the mid-level features, which are more transferable yet under-explored in main stream FSL work. To boost the discriminability of mid-level features, we propose a residual-prediction task to encourage mid-level features to learn discriminative information of each sample. Notably, such mechanism also benefits the in-domain FSL. Therefore, we provide two types of features for both distant- and in-domain FSL respectively, under the same training framework. Experiments under both settings on six public datasets, including two challenging medical datasets, validate the rationale of the proposed method and demonstrate state-of-the-art performance.