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

Although object detection has reached a milestone thanks to the great successof deep learning, the scale variation is still the key challenge. Integratingmulti-level features is presented to alleviate the problems, like the classicFeature Pyramid Network (FPN) and its improvements. However, the specificallydesigned feature integration modules of these methods may not have the optimalarchitecture for feature fusion. Moreover, these models have fixedarchitectures and data flow paths, when fed with various samples. They cannotadjust and be compatible with each kind of data. To overcome the abovelimitations, we propose a Dynamic Sample-Individualized Connector (DSIC) formulti-scale object detection. It dynamically adjusts network connections to fitdifferent samples. In particular, DSIC consists of two components: Intra-scaleSelection Gate (ISG) and Cross-scale Selection Gate (CSG). ISG adaptivelyextracts multi-level features from backbone as the input of featureintegration. CSG automatically activate informative data flow paths based onthe multi-level features. Furthermore, these two components are bothplug-and-play and can be embedded in any backbone. Experimental resultsdemonstrate that the proposed method outperforms the state-of-the-arts.

DSIC: Dynamic Sample-Individualized Connector for Multi-Scale Object Detection