Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ho, Dion J. X., Rong, Gabriel Lee Jun, Shrivastava, Niharika, Abichandani, Harshavardhan, Ng, Pai Chet, Miao, Xiaoxiao
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2510.12468
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917012163264512
author Ho, Dion J. X.
Rong, Gabriel Lee Jun
Shrivastava, Niharika
Abichandani, Harshavardhan
Ng, Pai Chet
Miao, Xiaoxiao
author_facet Ho, Dion J. X.
Rong, Gabriel Lee Jun
Shrivastava, Niharika
Abichandani, Harshavardhan
Ng, Pai Chet
Miao, Xiaoxiao
contents We present MS-GAGA (Metric-Selective Guided Adversarial Generation Attack), a two-stage framework for crafting transferable and visually imperceptible adversarial examples against deepfake detectors in black-box settings. In Stage 1, a dual-stream attack module generates adversarial candidates: MNTD-PGD applies enhanced gradient calculations optimized for small perturbation budgets, while SG-PGD focuses perturbations on visually salient regions. This complementary design expands the adversarial search space and improves transferability across unseen models. In Stage 2, a metric-aware selection module evaluates candidates based on both their success against black-box models and their structural similarity (SSIM) to the original image. By jointly optimizing transferability and imperceptibility, MS-GAGA achieves up to 27% higher misclassification rates on unseen detectors compared to state-of-the-art attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2510_12468
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MS-GAGA: Metric-Selective Guided Adversarial Generation Attack
Ho, Dion J. X.
Rong, Gabriel Lee Jun
Shrivastava, Niharika
Abichandani, Harshavardhan
Ng, Pai Chet
Miao, Xiaoxiao
Computer Vision and Pattern Recognition
We present MS-GAGA (Metric-Selective Guided Adversarial Generation Attack), a two-stage framework for crafting transferable and visually imperceptible adversarial examples against deepfake detectors in black-box settings. In Stage 1, a dual-stream attack module generates adversarial candidates: MNTD-PGD applies enhanced gradient calculations optimized for small perturbation budgets, while SG-PGD focuses perturbations on visually salient regions. This complementary design expands the adversarial search space and improves transferability across unseen models. In Stage 2, a metric-aware selection module evaluates candidates based on both their success against black-box models and their structural similarity (SSIM) to the original image. By jointly optimizing transferability and imperceptibility, MS-GAGA achieves up to 27% higher misclassification rates on unseen detectors compared to state-of-the-art attacks.
title MS-GAGA: Metric-Selective Guided Adversarial Generation Attack
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.12468