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

Fine-grained visual classification is a challenging task that recognizes thesub-classes belonging to the same meta-class. Large inter-class similarity andintra-class variance is the main challenge of this task. Most exiting methodstry to solve this problem by designing complex model structures to explore moreminute and discriminative regions. In this paper, we argue that miningmulti-regional multi-grained features is precisely the key to this task.Specifically, we introduce a new loss function, termed top-down spatialattention loss (TDSA-Loss), which contains a multi-stage channel constrainedmodule and a top-down spatial attention module. The multi-stage channelconstrained module aims to make the feature channels in different stagescategory-aligned. Meanwhile, the top-down spatial attention module uses theattention map generated by high-level aligned feature channels to makemiddle-level aligned feature channels to focus on particular regions. Finally,we can obtain multiple discriminative regions on high-level feature channelsand obtain multiple more minute regions within these discriminative regions onmiddle-level feature channels. In summary, we obtain multi-regionalmulti-grained features. Experimental results over four widely used fine-grainedimage classification datasets demonstrate the effectiveness of the proposedmethod. Ablative studies further show the superiority of two modules in theproposed method. Codes are available at:https://github.com/dongliangchang/Top-Down-Spatial-Attention-Loss.

Fine-Grained Visual Classification via Simultaneously Learning of Multi-regional Multi-grained Features