ProjectMuseumsBlick – The Museum Gaze

Basic data

The Museum Gaze
01/03/2022 to 28/02/2025
Abstract / short description:
Research context: In the art museum attention is focused through the eye. In order to understand viewing
behaviour, it is crucial to analyse the visitors’ mostly unconscious eye movements in relation to their self-reports.
Recent technical advances have made it easier to record eye movements in naturalistic settings.
However, to date, most eye tracking studies on art perception have been conducted in laboratories with twodimensional reproductions. Mobile eye tracking (MET) studies in museums have not yet realized their full
potential due to small sample sizes, constrained viewing conditions and laborious data analysis.

Hypotheses/research questions/objectives: The project uses MET to investigate the beholding of artworks in
the museum. The goals are 1) to study the relational gaze, i.e. how visitors alternate between looking at
single artworks, written information and other objects of interest; 2) to study the medium-specific gaze, i.e.
how visitors behave when beholding sculptures in contrast to paintings; and 3) to empower MET for the art
museum, i.e. to develop new open source solutions for automatic data annotation and 3D mapping including
indoor position tracking.

Approach/methods: The project intertwines technological developments (Tübingen) and their application in
four museum studies (Vienna) with increasing technical challenges. Every study will be bipartite to collect
data before and after executing display changes. The core method of MET is complemented by self-reported
visual and verbal data in a subjective mapping and a questionnaire registering reactions to artworks and
exhibition displays.

Level of originality/innovation: By combining art history, museum studies and computer science in an
innovative manner, we achieve a deeper level of analysis of the actual viewing behaviour of museum visitors.
The results will have significant effects 1) for art history, generating new knowledge on art perception and
delivering for the first time empirical data on the perception of sculptures; 2) for museum studies, providing a
profound empirical base on how visitor view and understand art in different exhibition displays and delivering
new tools to answer curatorial queries; 3) for computer science, advancing eye tracking methodology and
computer vision methods to egocentric videos for ecologically valid investigations of human behaviour.
eye tracking
machine learning
maschinelles Lernen
Human-Computer Interaction

Involved staff


Department of Informatics
Faculty of Science

Local organizational units

Department of Informatics
Faculty of Science
University of Tübingen


Bonn, Nordrhein-Westfalen, Germany

will be deleted permanently. This cannot be undone.