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

In this paper, we study the sensitivity of CNN outputs with respect to imagetransformations and noise in the area of fine-grained recognition. Inparticular, we answer the following questions (1) how sensitive are CNNs withrespect to image transformations encountered during wild image capture?; (2)how can we predict CNN sensitivity?; and (3) can we increase the robustness ofCNNs with respect to image degradations? To answer the first question, weprovide an extensive empirical sensitivity analysis of commonly used CNNarchitectures (AlexNet, VGG19, GoogleNet) across various types of imagedegradations. This allows for predicting CNN performance for new domainscomprised by images of lower quality or captured from a different viewpoint. Wealso show how the sensitivity of CNN outputs can be predicted for singleimages. Furthermore, we demonstrate that input layer dropout or pre-filteringduring test time only reduces CNN sensitivity for high levels of degradation.  Experiments for fine-grained recognition tasks reveal that VGG19 is morerobust to severe image degradations than AlexNet and GoogleNet. However, smallintensity noise can lead to dramatic changes in CNN performance even for VGG19.

Fine-grained Recognition in the Noisy Wild: Sensitivity Analysis of Convolutional Neural Networks Approaches