How people see
your photographs

See where attention gathers, and how the eye likely moves through the frame. Built for photographers, artists, and anyone curious about the invisible structure within an image.

JPEG or PNG · one image at a time Large images are resized automatically
Advanced settings
Model
Scanpath
4
0.40
0.08
0.22
0.07
Centre-first
0.30
0.21
Heatmap
0.75
0.50
0.80

Heatmap

Shows where attention is most likely to gather in the frame.

Based on human attention data.

Example heatmap

Scanpath

Shows the likely sequence of eye movements through the image.

Based on real eye-tracking research.

Example scanpath

Privacy

Images are analysed only for this session and are not stored.

Support Salience

If this tool helped you read your image differently, you can support the project and its continued development.

About

Salience is a project by photographer Martin Kocjan exploring the meeting point between composition and human attention, and the way vision-science models can help reveal it.

Get in touch

kocjanfineart@gmail.com

Research Credits

Salience draws on the following saliency and scanpath models:

  • UNISAL Droste, R., Jiao, J., & Noble, J. A. (2020). Unified Image and Video Saliency Modeling. In Proceedings of the 16th European Conference on Computer Vision (ECCV).
  • DeepGaze IIE Linardos, A., Kümmerer, M., Press, O., & Bethge, M. (2021). Calibrated prediction in and out-of-domain for state-of-the-art saliency modeling.
  • DeepGaze III Kümmerer, M., Bethge, M., & Wallis, T. S. A. (2022). DeepGaze III: Modeling free-viewing human scanpaths with deep learning. Journal of Vision.