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.
Cite
@article{arxiv.2506.09777,
title = {Inverting Black-Box Face Recognition Systems via Zero-Order Optimization in Eigenface Space},
author = {Anton Razzhigaev and Matvey Mikhalchuk and Klim Kireev and Igor Udovichenko and Andrey Kuznetsov and Aleksandr Petiushko},
journal= {arXiv preprint arXiv:2506.09777},
year = {2025}
}