Although human faces share similar features globally, static and dynamic facial details are highly individual. To investigate their impact on perception and cognitive processing of faces in neuroscientific research, the ability to virtually manipulate and synthesize such features is highly valuable.
At Zuse Institut Berlin, we are developing methods to establish 3D Statistical Face Models (SFM). SFMs allow to synthesize realistic faces by parametric manipulation of global and local facial features learned from 3D face data. The goal of this project is to improve our fully automatic face matching method [Grewe et al. 2018] in order to establish a SFM from a large heterogeneous database.

Please see the attachment for any details.

Research Group: 
Evaluation and Comparison of Linear and Non-linear Approaches for the Analysis and Synthesis of Facial Motion