Saliency Revisited: Analysis of Mouse Movements versus Fixations

TL;DR

This paper critically evaluates the use of mouse tracking data for visual saliency prediction, comparing it to eye tracking, and introduces a contextual evaluation methodology to analyze model performance across different visual regions.

Contribution

It provides a comprehensive analysis of mouse versus eye tracking data for saliency models and proposes a new evaluation scheme incorporating contextual information.

Findings

Mouse tracking data shows lower inter-participant consistency than eye tracking.

Mouse tracking data does not fully align with eye tracking, especially in specific regions.

Mouse tracking data can train models but is less reliable for model evaluation.

Abstract

This paper revisits visual saliency prediction by evaluating the recent advancements in this field such as crowd-sourced mouse tracking-based databases and contextual annotations. We pursue a critical and quantitative approach towards some of the new challenges including the quality of mouse tracking versus eye tracking for model training and evaluation. We extend quantitative evaluation of models in order to incorporate contextual information by proposing an evaluation methodology that allows accounting for contextual factors such as text, faces, and object attributes. The proposed contextual evaluation scheme facilitates detailed analysis of models and helps identify their pros and cons. Through several experiments, we find that (1) mouse tracking data has lower inter-participant visual congruency and higher dispersion, compared to the eye tracking data, (2) mouse tracking data does not totally agree with eye tracking in general and in terms of different contextual regions in specific, and (3) mouse tracking data leads to acceptable results in training current existing models, and (4) mouse tracking data is less reliable for model selection and evaluation. The contextual evaluation also reveals that, among the studied models, there is no single model that performs best on all the tested annotations.