What am I Searching for: Zero-shot Target Identity Inference in Visual Search

TL;DR

This paper introduces InferNet, a zero-shot model that infers search targets from eye movement error fixations, outperforming other models without needing object-specific training, thus advancing understanding of intention decoding from visual search behavior.

Contribution

The paper presents a novel zero-shot neural network model, InferNet, that infers search targets from eye movements without prior object-specific training, demonstrating improved accuracy.

Findings

InferNet outperforms baseline models in target inference accuracy.

Error fixations contain sufficient information to decode search intent.

The model generalizes across different search scenarios without additional training.

Abstract

Can we infer intentions from a person's actions? As an example problem, here we consider how to decipher what a person is searching for by decoding their eye movement behavior. We conducted two psychophysics experiments where we monitored eye movements while subjects searched for a target object. We defined the fixations falling on non-target objects as "error fixations". Using those error fixations, we developed a model (InferNet) to infer what the target was. InferNet uses a pre-trained convolutional neural network to extract features from the error fixations and computes a similarity map between the error fixations and all locations across the search image. The model consolidates the similarity maps across layers and integrates these maps across all error fixations. InferNet successfully identifies the subject's goal and outperforms competitive null models, even without any object-specific training on the inference task.