paper-with-me

홈 › Papers

Understanding Semantic Perturbations on In-Processing Generative Image Watermarks

2026-03-29 · Anirudh Nakra, Min Wu arxiv

The widespread deployment of high-fidelity generative models has intensified the need for reliable mechanisms for provenance and content authentication. In-processing watermarking, embedding a signature into the generative model's synthesis procedure, has been advocated as a solution and is often reported to be robust to standard post-processing (such as geometric transforms and filtering). Yet robustness to semantic manipulations that alter high-level scene content while maintaining reasonable visual quality is not well studied or understood. We introduce a simple, multi-stage framework for systematically stress-testing in-processing generative watermarks under semantic drift. The framework utilizes off-the-shelf models for object detection, mask generation, and semantically guided inpainting or regeneration to produce controlled, meaning-altering edits with minimal perceptual degradation. Based on extensive experiments on representative schemes, we find that robustness varies significantly with the degree of semantic entanglement: methods by which watermarks remain detectable under a broad suite of conventional perturbations can fail under semantic edits, with watermark detectability in many cases dropping to near zero while image quality remains high. Overall, our results reveal a critical gap in current watermarking evaluations and suggest that watermark designs and benchmarking must explicitly account for robustness against semantic manipulation.

📄 PDF Abstract BibTeX arXiv:2603.27513

Code (0)

등록된 구현이 없습니다.

Tasks

Object Detection

Similar Papers 제목 키워드 기반

A Generative Adversarial Approach to Adversarial Attacks Guided by Contrastive Language-Image Pre-trained Model

2025-11-03 · Sampriti Soor, Alik Pramanick, Jothiprakash K, Arijit Sur arxiv

The rapid growth of deep learning has brought about powerful models that can handle various tasks, like identifying images and understanding language. However, adversarial attacks, an unnoticed alteration, can deceive mo…

Adversarial Attack

CLINE: Contrastive Learning with Semantic Negative Examples for Natural Language Understanding

2021-07-01 · ACL 2021 5 · Dong Wang, Ning Ding, Piji Li, Hai-Tao Zheng

Despite pre-trained language models have proven useful for learning high-quality semantic representations, these models are still vulnerable to simple perturbations. Recent works aimed to improve the robustness of pre-tr…

Contrastive LearningNatural Language UnderstandingReading ComprehensionSentence+1

Generative Adversarial Perturbations

2017-12-06 · CVPR 2018 6 · Omid Poursaeed, Isay Katsman, Bicheng Gao, Serge Belongie

In this paper, we propose novel generative models for creating adversarial examples, slightly perturbed images resembling natural images but maliciously crafted to fool pre-trained models. We present trainable deep neura…

General ClassificationSemantic Segmentation

Semantically Adversarial Learnable Filters

2020-08-13 · Ali Shahin Shamsabadi, Changjae Oh, Andrea Cavallaro

We present an adversarial framework to craft perturbations that mislead classifiers by accounting for the image content and the semantics of the labels. The proposed framework combines a structure loss and a semantic adv…

Learning to Generate Training Datasets for Robust Semantic Segmentation

2023-08-01 · Marwane Hariat, Olivier Laurent, Rémi Kazmierczak, Shihao Zhang 외

Semantic segmentation methods have advanced significantly. Still, their robustness to real-world perturbations and object types not seen during training remains a challenge, particularly in safety-critical applications. …

Generative Adversarial NetworkSegmentationSemantic Segmentation