Saved in:
Bibliographic Details
Main Authors: Nakra, Anirudh, Wu, Min
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.27513
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917365918203904
author Nakra, Anirudh
Wu, Min
author_facet Nakra, Anirudh
Wu, Min
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_27513
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Understanding Semantic Perturbations on In-Processing Generative Image Watermarks
Nakra, Anirudh
Wu, Min
Computer Vision and Pattern Recognition
Artificial Intelligence
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.
title Understanding Semantic Perturbations on In-Processing Generative Image Watermarks
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2603.27513