Determine which robot gaze motions people perceive as more natural

Determine which of the Realistic Gaze Transformer, Eyes-Only, Head-Only, and Eye–Head Coordination controllers people perceive as producing more natural gaze motion through controlled observer ratings.

Background

The qualitative comparison presents one selected example of each controller's behavior, but the study collected no controlled observer ratings. The reported MSE and illustrative frames therefore do not establish how human observers judge the naturalness of the different motion-generation strategies.

A controlled observer study is needed to resolve which controller's gaze motion is perceived as more natural. This is distinct from the kinematic comparison because robot-space agreement does not directly measure human perception.