UEyes: An Eye-Tracking Dataset across User Interface Types

TL;DR

The paper presents UEyes, a high-fidelity eye-tracking dataset capturing how users view four common UI types, enabling direct comparison of bottom-up and top-down saliency cues. It documents a rigorous in-lab data collection with 62 participants viewing 1,980 UI screenshots (495 per type) and provides ~20K saliency maps and scanpaths, along with metadata. The work highlights a consistent top-left bias, stronger attention to text, and distinct viewing strategies, offering insights for improving cross-domain saliency models and personalized UI design analysis. By releasing open-science resources and outlining future directions, the study has practical impact for UI researchers and designers seeking accurate gaze-prediction tools across diverse interfaces.

Abstract

Different types of user interfaces differ significantly in the number of elements and how they are displayed. To examine how such differences affect the way users look at UIs, we collected and analyzed a large eye-tracking-based dataset, UEyes (62 participants, 1,980 UI screenshots, near 20K eye movement sequences), covering four major UI types: webpage, desktop UI, mobile UI, and poster. Furthermore, we analyze and discuss the differences in important factors, such as color, location, and gaze direction across UI types, individual viewing strategies and potential future directions. This position paper is a derivative of our recent paper with a particular focus on the UEyes dataset.

Figures (2)

• Figure 1: Examples of saliency maps and scanpaths in the UEyes dataset.
• Figure 2: Different users have different viewing strategies on user interfaces. Image-oriented users often look at images before text, while text-oriented users prefer the opposite.