Gravitational Models Explain Shifts on Human Visual Attention

TL;DR

This paper introduces a gravitational model for human visual attention shifts, which predicts attentional movements more accurately than traditional winner-take-all models by considering features as attractors.

Contribution

The paper proposes a novel gravitational model that describes attentional shifts without relying solely on a centralized saliency map, improving prediction accuracy.

Findings

The gravitational model outperforms winner-take-all in predicting attention shifts.

Quantitative results on large datasets validate the model's effectiveness.

The model suggests a decentralized approach to visual saliency.

Abstract

Visual attention refers to the human brain's ability to select relevant sensory information for preferential processing, improving performance in visual and cognitive tasks. It proceeds in two phases. One in which visual feature maps are acquired and processed in parallel. Another where the information from these maps is merged in order to select a single location to be attended for further and more complex computations and reasoning. Its computational description is challenging, especially if the temporal dynamics of the process are taken into account. Numerous methods to estimate saliency have been proposed in the last three decades. They achieve almost perfect performance in estimating saliency at the pixel level, but the way they generate shifts in visual attention fully depends on winner-take-all (WTA) circuitry. WTA is implemented} by the biological hardware in order to select a location with maximum saliency, towards which to direct overt attention. In this paper we propose a gravitational model (GRAV) to describe the attentional shifts. Every single feature acts as an attractor and {the shifts are the result of the joint effects of the attractors. In the current framework, the assumption of a single, centralized saliency map is no longer necessary, though still plausible. Quantitative results on two large image datasets show that this model predicts shifts more accurately than winner-take-all.