AI-generated faces show lower morphological diversity than real faces do
AI-generated faces show lower morphological diversity than real faces do

Summary/Abstract: Some recent studies suggest that artificial intelligence can create realistic human facessubjectively unrecognizable from faces of real people. We have compared static facial photographs of 197real men with a sample of 200 male faces generated by artificial intelligence to test whether they converge inbasic morphological characteristic such as shape variation and bilateral asymmetry. Both datasets depictedstandardized faces of European men with a neutral expression. Then we used geometric morphometricsto investigate their facial morphology and calculate the measures of shape variation and asymmetry. Wefound that the natural faces of real individuals were more variable in their facial shape than the artificiallygenerated faces were. Moreover, the artificially synthesized faces showed lower levels of facial asymmetrythan the control group. Despite the rapid development of generative adversarial networks, natural faces arethus still statistically distinguishable from the artificial ones by objective measurements. We recommendthe researchers in face perception, that aim to use artificially generated faces as ecologically valid stimuli,to check whether their stimuli morphological variance is comparable with that of natural faces in a targetpopulation.

• Details
• Contents