Few-shot learning with unsupervised part discovery and part-aligned similarity. (January 2023)

• Record Type: Journal Article
• Authors: Chen, Wentao; Zhang, Zhang; Wang, Wei; Wang, Liang; Wang, Zilei; Tan, Tieniu
• Is Part Of: Pattern recognition. Volume 133(2023)
• Publication Date: 2023-01
• Subjects: Few-shot learning -- Self-supervised learning -- Part discovery network -- Part-aligned similarity; 00–01 -- 99-00; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2022.108986
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We propose a novel unsupervised Part Discovery Network, which can learn discriminative and transferable part representations from unlabeled images for few-shot learning. We propose Part-Aligned Similarity, which measures image similarities based on discriminative and aligned parts via partweighting and part-alignment mechanisms. We conduct extensive experiments on five few-shot learning benchmarks. The experimental results demonstrate that the proposed approach outperforms previous unsupervised methods by a large margin and achieves comparable performance with supervised methods. Abstract: Few-shot learning aims to recognize novel concepts with only a few examples. To this end, previous studies resort to acquiring a strong inductive bias via meta-learning on a group of similar tasks, which however needs a large labeled base dataset to sample training 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 unsupervised Part Discovery Network (PDN) to learn transferable representations from unlabeled images, which automatically selects the most discriminative part from an input image and then maximizes its similarities to the global view of the input and other neighbors with similar semantics. To better leverage the learned representations for few-shot learning, we further propose Part-Aligned Similarity (PAS), the key of which is to measure image similarities based on a set of discriminative and aligned parts. We conduct extensive studies on five popular few-shot learning datasets to evaluate our approach. The experimental results show that our approach outperforms previous unsupervised methods by a large margin and is even comparable with state-of-the-art supervised methods.