Understanding and mapping pleasure, arousal and dominance social signals to robot-avatar behavior

摘要

We present an analysis of the pleasure, arousal, and dominance social signals inferred from people faces, and how, despite their noisy nature, these can be used to drive a model of theory-based interventions for a robot-avatar agent in a working space. The analysis let emerge clearly the need of data pre-filtering and per-user calibration. The proposed post processing method helps quantifying the parameters needed to control the frequency of intervention of the agent; still leaving the experimenter with a run-time adjustable global control of its sensitivity.