Mimicking others’ facial expressions is believed to be important in making virtual humans as more natural and believable. As result of an empirical study conducted with a virtual human a large face repertoire of about 6000 faces arranged in Pleasure Arousal Dominance (PAD-) space with respect to two dominance values (dominant vs. submissive) was obtained. Each face in the face repertoire consists of different intensities of the virtual human’s facial muscle actions called Action Units (AUs), modeled following the Facial Action Coding System (FACS). Using this face repertoire an approach towards realizing facial

mimicry for a virtual human is topic of this paper. A preliminary evaluation of this first approach is realized with the basic emotions Happy and Angry.