Saved in:
Bibliographic Details
Main Authors: Yang, Pei, Liu, Yepeng, Peng, Kelly, Gao, Yuan, Song, Yiren
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.01314
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915646882709504
author Yang, Pei
Liu, Yepeng
Peng, Kelly
Gao, Yuan
Song, Yiren
author_facet Yang, Pei
Liu, Yepeng
Peng, Kelly
Gao, Yuan
Song, Yiren
contents In the digital economy era, digital watermarking serves as a critical basis for ownership proof of massive replicable content, including AI-generated and other virtual assets. Designing robust watermarks capable of withstanding various attacks and processing operations is even more paramount. We introduce TokenPure, a novel Diffusion Transformer-based framework designed for effective and consistent watermark removal. TokenPure solves the trade-off between thorough watermark destruction and content consistency by leveraging token-based conditional reconstruction. It reframes the task as conditional generation, entirely bypassing the initial watermark-carrying noise. We achieve this by decomposing the watermarked image into two complementary token sets: visual tokens for texture and structural tokens for geometry. These tokens jointly condition the diffusion process, enabling the framework to synthesize watermark-free images with fine-grained consistency and structural integrity. Comprehensive experiments show that TokenPure achieves state-of-the-art watermark removal and reconstruction fidelity, substantially outperforming existing baselines in both perceptual quality and consistency.
format Preprint
id arxiv_https___arxiv_org_abs_2512_01314
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TokenPure: Watermark Removal through Tokenized Appearance and Structural Guidance
Yang, Pei
Liu, Yepeng
Peng, Kelly
Gao, Yuan
Song, Yiren
Computer Vision and Pattern Recognition
In the digital economy era, digital watermarking serves as a critical basis for ownership proof of massive replicable content, including AI-generated and other virtual assets. Designing robust watermarks capable of withstanding various attacks and processing operations is even more paramount. We introduce TokenPure, a novel Diffusion Transformer-based framework designed for effective and consistent watermark removal. TokenPure solves the trade-off between thorough watermark destruction and content consistency by leveraging token-based conditional reconstruction. It reframes the task as conditional generation, entirely bypassing the initial watermark-carrying noise. We achieve this by decomposing the watermarked image into two complementary token sets: visual tokens for texture and structural tokens for geometry. These tokens jointly condition the diffusion process, enabling the framework to synthesize watermark-free images with fine-grained consistency and structural integrity. Comprehensive experiments show that TokenPure achieves state-of-the-art watermark removal and reconstruction fidelity, substantially outperforming existing baselines in both perceptual quality and consistency.
title TokenPure: Watermark Removal through Tokenized Appearance and Structural Guidance
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.01314