1 paper · 1 filter
Byunghoon Oh, Sunghwan Park, Jaewoo Lee
Unlearnable examples keep publicly shared photos from being learned by unauthorized face-recognition models. An imperceptible perturbation, added before sharing, makes any model tr…