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

The class-agnostic counting (CAC) task has recently been proposed to solvethe problem of counting all objects of an arbitrary class with severalexemplars given in the input image. To address this challenging task, existingleading methods all resort to density map regression, which renders themimpractical for downstream tasks that require object locations and restrictstheir ability to well explore the scale information of exemplars forsupervision. To address the limitations, we propose a novel localization-basedCAC approach, termed Scale-modulated Query and Localization Network (SQLNet).It fully explores the scales of exemplars in both the query and localizationstages and achieves effective counting by accurately locating each object andpredicting its approximate size. Specifically, during the query stage, richdiscriminative representations of the target class are acquired by theHierarchical Exemplars Collaborative Enhancement (HECE) module from the fewexemplars through multi-scale exemplar cooperation with equifrequent sizeprompt embedding. These representations are then fed into the Exemplars-UnifiedQuery Correlation (EUQC) module to interact with the query features in aunified manner and produce the correlated query tensor. In the localizationstage, the Scale-aware Multi-head Localization (SAML) module utilizes the querytensor to predict the confidence, location, and size of each potential object.Moreover, a scale-aware localization loss is introduced, which exploitsflexible location associations and exemplar scales for supervision to optimizethe model performance. Extensive experiments demonstrate that SQLNetoutperforms state-of-the-art methods on popular CAC benchmarks, achievingexcellent performance not only in counting accuracy but also in localizationand bounding box generation. Our codes will be available athttps://github.com/HCPLab-SYSU/SQLNet

SQLNet: Scale-Modulated Query and Localization Network for Few-Shot Class-Agnostic Counting