CReA Lab

Methods and Software

The lab has analysed eye movements in front of artworks since 2006. Since 2024 it also develops the methods itself: published algorithms, open-source software, computer vision for eye-tracking glasses, and courses that teach the tools to art historians. The work is coordinated by Xingyu Long; in questions of modelling and statistics the lab cooperates with József Arató (Budapest University of Technology and Economics).

Software

  • HeatMatch: orientation-aware heat maps of dense saccade patterns and a similarity measure between viewers, groups and artworks (Long et al. 2026). Code on GitHub, package on PyPI.
  • creapy: desktop application and Python library for the lab's standard eye-tracking analyses, from heat maps and scanpaths to areas of interest, time courses and similarity. It reads EyeLink and Tobii Pro Lab data and is made for art historians without programming experience (Long & Rosenberg 2026). Code on GitHub, package on PyPI.
  • Compolines: a tablet tool for drawing compositional lines; the drawings are compared with each other and with eye movements on the same painting (Long et al. 2027).

Teaching

Support

Lab members and students get support from the planning of a study (analysis plans, pre-registrations, recording settings) to the analysis of existing data (cleaning, measures, figures, statistics).

Publications

  • Long, X., Hsu, C.-S., Balázs, R., Salomvári, Á., Kury, S., & Rosenberg, R. (2027). Unsichere Blicke, feste Striche: Kompositionszeichnen als Verhaltenserhebung für empirische Bildwissenschaft. DHd 2027: Mind the gap! – Knowledge, Uncertainty and Responsibility, Marburg, 1–5 March 2027 (accepted).
  • Long, X., Arató, J., Kury, S., Miscenà, A., & Rosenberg, R. (2026). HeatMatch: Orientation-Aware Visualization and Comparison of Very Dense Saccade Patterns. Proceedings of the ACM on Computer Graphics and Interactive Techniques, 9(2), Article 18. https://doi.org/10.1145/3803539
  • Long, X., & Rosenberg, R. (2026). Perspektiven für ein anfängerfreundliches Tool zur Eye-Tracking-Datenvisualisierung in der kognitiven Kunstgeschichte. In S. Schwandt, G. Viehhauser, T. Andrews, & T. Wallnig (Eds.), Book of Abstracts – DHd 2026: Not Only Text, Not Only Data (pp. 606–607). Zenodo. https://doi.org/10.5281/zenodo.18703017
Saccade patterns of five viewers (columns) on three still lifes (rows), computed with HeatMatch; the last column aggregates all 46 viewers.
CReA Lab