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

Deep convolutional neural networks (CNN) always depend on wider receptivefield (RF) and more complex non-linearity to achieve state-of-the-artperformance, while suffering the increased difficult to interpret how relevantpatches contribute the final prediction. In this paper, we construct aninterpretable AnchorNet equipped with our carefully designed RFs and linearlyspatial aggregation to provide patch-wise interpretability of the input mediameanwhile localizing multi-scale informative patches only supervised onmedia-level labels without any extra bounding box annotations. Visualization oflocalized informative image and text patches show the superior multi-scalelocalization capability of AnchorNet. We further use localized patches fordownstream classification tasks across widely applied networks. Experimentalresults demonstrate that replacing the original inputs with their patches forclassification can get a clear inference acceleration with only tinyperformance degradation, which proves that localized patches can indeed retainthe most semantics and evidences of the original inputs.

Localizing Interpretable Multi-scale informative Patches Derived from Media Classification Task