According to the paper linked below (which has Bastioni as one of the authors), both the FACS system and MPEG-4 had an influence on the development of MakeHuman's expression pose model. In the last page they write "The whole process of translation into targets has carefully followed MPEG-4's descriptions of FAPs". This explains why most of the parameters in the expression pose model are named similarly to those in MPEG-4.
https://visagetechnologies.com/uploads/ ... erview.pdf