Few-Shot Learning with Part Discovery and Augmentation from Unlabeled Images

TL;DR

This paper introduces a novel part-based self-supervised learning approach combined with part augmentation to improve few-shot image classification, achieving results comparable to supervised methods on standard benchmarks.

Contribution

It proposes a new part-based self-supervised representation learning scheme and a part augmentation strategy to enhance transferability and reduce overfitting in few-shot learning.

Findings

Outperforms previous unsupervised methods by 7.74% and 9.24% on miniImageNet.

Achieves results comparable to state-of-the-art supervised methods.

Demonstrates effectiveness of part-based self-supervision and augmentation in few-shot tasks.

Abstract

Few-shot learning is a challenging task since only few instances are given for recognizing an unseen class. One way to alleviate this problem is to acquire a strong inductive bias via meta-learning on similar tasks. In this paper, we show that such inductive bias can be learned from a flat collection of unlabeled images, and instantiated as transferable representations among seen and unseen classes. Specifically, we propose a novel part-based self-supervised representation learning scheme to learn transferable representations by maximizing the similarity of an image to its discriminative part. To mitigate the overfitting in few-shot classification caused by data scarcity, we further propose a part augmentation strategy by retrieving extra images from a base dataset. We conduct systematic studies on miniImageNet and tieredImageNet benchmarks. Remarkably, our method yields impressive results, outperforming the previous best unsupervised methods by 7.74% and 9.24% under 5-way 1-shot and 5-way 5-shot settings, which are comparable with state-of-the-art supervised methods.