Guardado en:
Detalles Bibliográficos
Autores principales: Abecidan, Rony, Itier, Vincent, Boulanger, Jérémie, Bas, Patrick, Pevný, Tomáš
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2605.21523
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913150871273472
author Abecidan, Rony
Itier, Vincent
Boulanger, Jérémie
Bas, Patrick
Pevný, Tomáš
author_facet Abecidan, Rony
Itier, Vincent
Boulanger, Jérémie
Bas, Patrick
Pevný, Tomáš
contents Steganalysis models excel on benchmark datasets but struggle in the wild when analyzed images are produced by a processing pipeline unseen during training. This problem known as Cover Source Mismatch (CSM) is particularly hard in realistic settings where practitioners (1) have access to only a small, unlabeled dataset, (2) are unsure of the processing techniques applied to these images, and (3) lack information on the proportion of covers and stegos in that set. To answer this challenge, we introduce TADA (Target Alignment through Data Adaptation), a framework learning to emulate the unknown processing pipeline from a small unlabeled target set. This architecture is trained with a loss combining residual covariance alignment, residual distribution matching, and a $\ell^2$ loss constraining the emulator to produce realistic images. Across toy and operational targets, TADA yields substantial gains in robustness to CSM and improves operational generalization compared to strong holistic and atomistic baselines. Additional resources are available at this link: https://github.com/RonyAbecidan/TADA
format Preprint
id arxiv_https___arxiv_org_abs_2605_21523
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Tackle CSM in JPEG Steganalysis with Data Adaptation
Abecidan, Rony
Itier, Vincent
Boulanger, Jérémie
Bas, Patrick
Pevný, Tomáš
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Multimedia
Signal Processing
Steganalysis models excel on benchmark datasets but struggle in the wild when analyzed images are produced by a processing pipeline unseen during training. This problem known as Cover Source Mismatch (CSM) is particularly hard in realistic settings where practitioners (1) have access to only a small, unlabeled dataset, (2) are unsure of the processing techniques applied to these images, and (3) lack information on the proportion of covers and stegos in that set. To answer this challenge, we introduce TADA (Target Alignment through Data Adaptation), a framework learning to emulate the unknown processing pipeline from a small unlabeled target set. This architecture is trained with a loss combining residual covariance alignment, residual distribution matching, and a $\ell^2$ loss constraining the emulator to produce realistic images. Across toy and operational targets, TADA yields substantial gains in robustness to CSM and improves operational generalization compared to strong holistic and atomistic baselines. Additional resources are available at this link: https://github.com/RonyAbecidan/TADA
title Tackle CSM in JPEG Steganalysis with Data Adaptation
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Multimedia
Signal Processing
url https://arxiv.org/abs/2605.21523