English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Few-shot learning (FSL) aims to learn novel visual categories from very fewsamples, which is a challenging problem in real-world applications. Manymethods of few-shot classification work well on general images to learn globalrepresentation. However, they can not deal with fine-grained categories well atthe same time due to a lack of subtle and local information. We argue thatlocalization is an efficient approach because it directly provides thediscriminative regions, which is critical for both general classification andfine-grained classification in a low data regime. In this paper, we propose aSelf-Attention Based Complementary Module (SAC Module) to fulfill theweakly-supervised object localization, and more importantly produce theactivated masks for selecting discriminative deep descriptors for few-shotclassification. Based on each selected deep descriptor, Semantic AlignmentModule (SAM) calculates the semantic alignment distance between the query andsupport images to boost classification performance. Extensive experiments showour method outperforms the state-of-the-art methods on benchmark datasets undervarious settings, especially on the fine-grained few-shot tasks. Besides, ourmethod achieves superior performance over previous methods when training themodel on miniImageNet and evaluating it on the different datasets,demonstrating its superior generalization capacity. Extra visualization showsthe proposed method can localize the key objects more interval.

Weakly-supervised Object Localization for Few-shot Learning and Fine-grained Few-shot Learning