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 Learning models like Convolutional Neural Networks (CNN) are powerfulimage classifiers, but what factors determine whether they attend to similarimage areas as humans do? While previous studies have focused on technologicalfactors, little is known about the role of factors that affect human attention.In the present study, we investigated how the tasks used to elicit humanattention maps interact with image characteristics in modulating the similaritybetween humans and CNN. We varied the intentionality of human tasks, rangingfrom spontaneous gaze during categorization over intentional gaze-pointing upto manual area selection. Moreover, we varied the type of image to becategorized, using either singular, salient objects, indoor scenes consistingof object arrangements, or landscapes without distinct objects defining thecategory. The human attention maps generated in this way were compared to theCNN attention maps revealed by explainable artificial intelligence (Grad-CAM).The influence of human tasks strongly depended on image type: For objects,human manual selection produced maps that were most similar to CNN, while thespecific eye movement task has little impact. For indoor scenes, spontaneousgaze produced the least similarity, while for landscapes, similarity wasequally low across all human tasks. To better understand these results, we alsocompared the different human attention maps to each other. Our resultshighlight the importance of taking human factors into account when comparingthe attention of humans and CNN.

Do humans and Convolutional Neural Networks attend to similar areas during scene classification: Effects of task and image type