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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Few-shot classification which aims to recognize unseen classes using verylimited samples has attracted more and more attention. Usually, it isformulated as a metric learning problem. The core issue of few-shotclassification is how to learn (1) consistent representations for images inboth support and query sets and (2) effective metric learning for imagesbetween support and query sets. In this paper, we show that the two challengescan be well modeled simultaneously via a unified Query-Support TransFormer(QSFormer) model. To be specific,the proposed QSFormer involves globalquery-support sample Transformer (sampleFormer) branch and local patchTransformer (patchFormer) learning branch. sampleFormer aims to capture thedependence of samples in support and query sets for image representation. Itadopts the Encoder, Decoder and Cross-Attention to respectively model theSupport, Query (image) representation and Metric learning for few-shotclassification task. Also, as a complementary to global learning branch, weadopt a local patch Transformer to extract structural representation for eachimage sample by capturing the long-range dependence of local image patches. Inaddition, a novel Cross-scale Interactive Feature Extractor (CIFE) is proposedto extract and fuse multi-scale CNN features as an effective backbone modulefor the proposed few-shot learning method. All modules are integrated into aunified framework and trained in an end-to-end manner. Extensive experiments onfour popular datasets demonstrate the effectiveness and superiority of theproposed QSFormer.

Few-Shot Learning Meets Transformer: Unified Query-Support Transformers for Few-Shot Classification