Continuous prediction of trait impressions

In this paper, we address perceived personality trait prediction problem from a novel perspective. First, in the course of generating ground-truth, we ask external observers to continuously provide ratings along multiple dimensions ranging from 0 to 100 along time, and we generate continuous annotations in space and time. Secondly, in addition to the widely used Big Five personality dimensions, we introduce four more dimensions which has the potential to gauge the reliability of the perceived social and trait judgements. Preliminary results demonstrate the viability of the proposed approach in the context of interactions between a human subject and virtual characters.