Classifying Head Movements to Separate Head-Gaze and Head Gestures as Distinct Modes of Input

Abstract

Head movement is widely used as a uniform type of input for human-computer interaction. However, there are fundamental differences between head movements coupled with gaze in support of our visual system, and head movements performed as gestural expression. Both Head-Gaze and Head Gestures are of utility for interaction but difer in their afordances. To facilitate the treatment of Head-Gaze and Head Gestures as separate types of input, we developed HeadBoost as a novel classifer, achieving high accuracy in classifying gaze-driven versus gestural head movement ( 1 -Score: 0.89). We demonstrate the utility of the classifer with three applications: gestural input while avoiding unintentional input by Head-Gaze; target selection with Head-Gaze while avoiding Midas Touch by head gestures; and switching of cursor control between Head-Gaze for fast positioning and Head Gesture for refnement. The classifcation of Head-Gaze and Head Gesture allows for seamless head-based interaction while avoiding false activation. CCS CONCEPTS• Human-centered computing → Gestural input; Virtual reality;

GazeSwitch: Automatic Eye-Head Mode Switching for Optimised Hands-Free Pointing

“…Besides hand-based input, head input has shown to be a promising input for disambiguation and refinement pointing for target selection, as the head affords hands-free fine control. In our previous work, we found that users have fine-grained over their head movement (∼0.3 degrees) [17]. Eye-head combination capitalises on the strengths of both modalities, with the eyes providing fast and precise input while head movements enable finer adjustments.…”

“…In our work, we build on the insights from BimodalGaze for automatic eye-head mode switching, and the potential of head movement classification from our previous work, HeadBoost [17]. The HeadBoost classifier addresses the challenge of correctly classifying between two fundamental types of head movements: gaze-driven head movement (Head-Gaze) and gestural head movement (Head Gesture).…”

“…To develop GazeSwitch, we first obtained a labelled dataset of eye and head movement data of participants as they performed cursor refinement tasks in a controlled study (detailed in Section 3.1). We closely followed the pipeline steps used to develop HeadBoost [17], including preprocessing and the initial steps for feature engineering (Section 3.2). With recursive feature selection, we obtained a classification rate above 120 Hz with a high classification performance of 0.91 𝐹 1 -Score (Section 3.3).…”

“…We followed the target design used in HeadBoost [17], featuring a diameter of 5.72 degrees, a transparent centre of 2.56 degrees (Figure 1a), and a black dot of 0.8 degrees in diameter (Figure 1c). The small size of the black dot was chosen to challenge gaze pointing, thereby encouraging participants to invoke head mode for refinement.…”

“…In this paper, we contribute GazeSwitch, an automatic ML-based approach for real-time switching between eye and head mode for optimised hands-free cursor control in mixed and extended reality. GazeSwitch leverages insights from HeadBoost [17], our previous work, which introduced a method for separating gaze-driven and gestural head movements using a machine learning approach. In evaluation, HeadBoost proved effective compared to BimodalGaze's threshold-based approach [37].…”