Beyond FID: Human Perceptual Judgments Reveal Systematic Blind Spots in GAN Face Evaluation

Simple item page
dc.contributor.authorNierula, Birgit
dc.contributor.authorMelnik, Anna
dc.contributor.authorBarthel, Florian
dc.contributor.authorBrama, Aileen
dc.contributor.authorHilsmann, Anna
dc.contributor.authorEisert, Peter
dc.contributor.authorNikulin, Vadim V.
dc.contributor.authorGaebler, Michael
dc.contributor.authorKlotzsche, Felix
dc.contributor.authorChen, Yonghao
dc.contributor.authorStephani, Tilman
dc.contributor.authorBosse, Sebastian
dc.contributor.editorMusialski, Przemyslaw
dc.contributor.editorLim, Isaak
dc.date.accessioned2026-04-20T08:01:34Z
dc.date.available2026-04-20T08:01:34Z
dc.date.issued2026
dc.description.abstractGenerative Adversarial Networks (GANs) can synthesize highly realistic facial images from random noise vectors. The Fréchet Inception Distance (FID) is widely used as a standard metric to automatically evaluate the quality of GAN-generated images. However, it remains unclear to what extent this statistical measure reflects human perceptual judgments, which ultimately define image realism in practical applications. To address this, we conducted a psychophysical study in which participants (n = 20) performed a two-alternative forced-choice task, assessing actual photographs and GAN-generated images as real or fake. We show that while FID provides a reliable global ordering of image quality, it systematically fails for localized semantic artifacts (e.g., eyewear and skin texture) that disproportionately affect human realness judgments. This demonstrates that FID and human perception are not merely noisy versions of the same signal, but that FID has systematic blind spots for localized semantic artifacts that disproportionately drive human realism judgments.
dc.description.sectionheadersFaces, Characters & Human Modeling
dc.description.seriesinformationEurographics 2026 - Short Papers
dc.identifier.doi10.2312/egs.20261007
dc.identifier.isbn978-3-03868-299-8
dc.identifier.issn2309-5059
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/egs20261007
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/egs20261007
dc.publisherThe Eurographics Association
dc.rightsCC-BY-4.0
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectModels of computation
dc.subjectInteractive computation
dc.subjectComputer Graphics
dc.titleBeyond FID: Human Perceptual Judgments Reveal Systematic Blind Spots in GAN Face Evaluation
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
egs20261007.pdf
Size:
1.43 MB
Format:
Adobe Portable Document Format
Download
Collections
EG 2026 - Short Papers