Inverting Black-Box Face Recognition Systems via Zero-Order Optimization in Eigenface Space

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Razzhigaev, Anton, Mikhalchuk, Matvey, Kireev, Klim, Udovichenko, Igor, Kuznetsov, Andrey, Petiushko, Aleksandr
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913889335115776
author Razzhigaev, Anton
Mikhalchuk, Matvey
Kireev, Klim
Udovichenko, Igor
Kuznetsov, Andrey
Petiushko, Aleksandr
author_facet Razzhigaev, Anton
Mikhalchuk, Matvey
Kireev, Klim
Udovichenko, Igor
Kuznetsov, Andrey
Petiushko, Aleksandr
contents Reconstructing facial images from black-box recognition models poses a significant privacy threat. While many methods require access to embeddings, we address the more challenging scenario of model inversion using only similarity scores. This paper introduces DarkerBB, a novel approach that reconstructs color faces by performing zero-order optimization within a PCA-derived eigenface space. Despite this highly limited information, experiments on LFW, AgeDB-30, and CFP-FP benchmarks demonstrate that DarkerBB achieves state-of-the-art verification accuracies in the similarity-only setting, with competitive query efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2506_09777
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Inverting Black-Box Face Recognition Systems via Zero-Order Optimization in Eigenface Space
Razzhigaev, Anton
Mikhalchuk, Matvey
Kireev, Klim
Udovichenko, Igor
Kuznetsov, Andrey
Petiushko, Aleksandr
Computer Vision and Pattern Recognition
Artificial Intelligence
Reconstructing facial images from black-box recognition models poses a significant privacy threat. While many methods require access to embeddings, we address the more challenging scenario of model inversion using only similarity scores. This paper introduces DarkerBB, a novel approach that reconstructs color faces by performing zero-order optimization within a PCA-derived eigenface space. Despite this highly limited information, experiments on LFW, AgeDB-30, and CFP-FP benchmarks demonstrate that DarkerBB achieves state-of-the-art verification accuracies in the similarity-only setting, with competitive query efficiency.
title Inverting Black-Box Face Recognition Systems via Zero-Order Optimization in Eigenface Space
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2506.09777