Ken Hawkins

Foveacast v3 · 3 s · FP16

What stands out in the first three seconds of viewing an interface

Open in Studio

What to provide

Attach a static PNG, JPEG or WebP screenshot up to 3 MB and 12 megapixels.

How it works

MSI-Net fine-tuned on the three-second UEyes viewing window estimates saliency. Studio preserves the screenshot’s aspect ratio and returns an overlay, concentration score and shares by section.

How to read the result

Compare a headline with the main button to see which draws more attention. The map estimates attention on a still image. Use a user study to observe actual gaze order or scrolling behavior.

Credits and licenses

Foveacast v3 · 3 sMIT

Model by Ken Hawkins based on MSI-Net (Kroner et al., 2020) and fine-tuned on UEyes.

UEyes: Understanding Visual Saliency across User Interface Types by Yue Jiang, Luis A. Leiva, Hamed Rezazadegan Tavakoli, Paul R. B. Houssel, Julia Kylmälä and Antti Oulasvirta (CHI 2023). Both attention models were fine-tuned on this dataset, released under CC BY 4.0.

Official documentation