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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Human visual attention is a complex phenomenon. A computational modeling ofthis phenomenon must take into account where people look in order to evaluatewhich are the salient locations (spatial distribution of the fixations), whenthey look in those locations to understand the temporal development of theexploration (temporal order of the fixations), and how they move from onelocation to another with respect to the dynamics of the scene and the mechanicsof the eyes (dynamics). State-of-the-art models focus on learning saliency mapsfrom human data, a process that only takes into account the spatial componentof the phenomenon and ignore its temporal and dynamical counterparts. In thiswork we focus on the evaluation methodology of models of human visualattention. We underline the limits of the current metrics for saliencyprediction and scanpath similarity, and we introduce a statistical measure forthe evaluation of the dynamics of the simulated eye movements. While deeplearning models achieve astonishing performance in saliency prediction, ouranalysis shows their limitations in capturing the dynamics of the process. Wefind that unsupervised gravitational models, despite of their simplicity,outperform all competitors. Finally, exploiting a crowd-sourcing platform, wepresent a study aimed at evaluating how strongly the scanpaths generated withthe unsupervised gravitational models appear plausible to naive and experthuman observers.

Toward Improving the Evaluation of Visual Attention Models: a Crowdsourcing Approach