SIGN: A Statistically-Informed Gaze Network for Gaze Time Prediction

TL;DR

SIGN introduces a novel deep learning model combining statistical insights to predict aggregate gaze times and underlying gaze patterns on images, outperforming existing benchmarks and aligning with empirical fixation data.

Contribution

The paper presents the first implementation of SIGN, a statistically-informed deep learning model that predicts gaze durations and patterns using CNNs and Visual Transformers.

Findings

SIGN significantly outperforms state-of-the-art benchmarks in gaze duration prediction.

SIGN produces plausible gaze pattern probability maps consistent with empirical data.

The model demonstrates versatility across datasets with aggregate and individual gaze data.

Abstract

We propose a first version of SIGN, a Statistically-Informed Gaze Network, to predict aggregate gaze times on images. We develop a foundational statistical model for which we derive a deep learning implementation involving CNNs and Visual Transformers, which enables the prediction of overall gaze times. The model enables us to derive from the aggregate gaze times the underlying gaze pattern as a probability map over all regions in the image, where each region's probability represents the likelihood of being gazed at across all possible scan-paths. We test SIGN's performance on AdGaze3500, a dataset of images of ads with aggregate gaze times, and on COCO-Search18, a dataset with individual-level fixation patterns collected during search. We demonstrate that SIGN (1) improves gaze duration prediction significantly over state-of-the-art deep learning benchmarks on both datasets, and (2) can deliver plausible gaze patterns that correspond to empirical fixation patterns in COCO-Search18. These results suggest that the first version of SIGN holds promise for gaze-time predictions and deserves further development.