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 have demonstrated high performances forfixation prediction in recent years. How they achieve this, however, is lessexplored and they remain to be black box models. Here, we attempt to shed lighton the internal structure of deep saliency models and study what features theyextract for fixation prediction. Specifically, we use a simple yet powerfularchitecture, consisting of only one CNN and a single resolution input,combined with a new loss function for pixel-wise fixation prediction duringfree viewing of natural scenes. We show that our simple method is on par orbetter than state-of-the-art complicated saliency models. Furthermore, wepropose a method, related to saliency model evaluation metrics, to visualizedeep models for fixation prediction. Our method reveals the innerrepresentations of deep models for fixation prediction and provides evidencethat saliency, as experienced by humans, is likely to involve high-levelsemantic knowledge in addition to low-level perceptual cues. Our results can beuseful to measure the gap between current saliency models and the humaninter-observer model and to build new models to close this gap.

What Catches the Eye? Visualizing and Understanding Deep Saliency Models