A face recognition model links two images of a person recorded on separate occasions when their embedding similarity exceeds an operating threshold. We consider making chosen identities unlinkable across separate occasions while the model remains in service for the rest of the population. Deleting their images and retr...
This paper proposes ScoreMix, a self-contained synthetic generation method to produce hard synthetic samples for recognition tasks by leveraging the score compositionality of diffusion models, and finds that mixing classes distant in the discriminator's embedding space yields larger gains, compared to selection based o...
The results indicate that the number of training identities has the largest effect on member/non-member separability, while backbone and loss head contribute far less, and that, on a same-domain held-out reference, the geometric membership signal decreases monotonically as more identities are added to training.
Ünsal Öztürk, Sébastien Marcel· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.