A pilot study on gaze and mouse data for user identification

The human gaze is increasingly used for different applications in interactive systems. It is used for explicit and implicit user input detection and is suitable for various purposes, such as activation commands, emotion detection, or user identification. Anyway, its reliability could not be good enough in some contexts or for some particular applications. In these cases, it is often integrated with other input sources to improve detection quality. This paper presents a protocol for gaze and mouse data acquisition, along with a possible usage for user identification. In particular, we show the experimental setup, the tasks the participants were asked to accomplish, and the data we acquired. We also propose a possible use case for these data and evaluate the results. We conclude by discussing how the data acquired with our protocol can be helpful in different applications and contexts, along with a plan for further development.