arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2503.02490cs.CV
收藏

深度鲁棒可逆水印

Deep Robust Reversible Watermarking

School of Computer Science, Beijing Institute of Technology(北京理工大学计算机学院);Artificial Intelligence Research Institute, Shenzhen MSU-BIT University(深圳北理莫斯科大学人工智能研究院);School of Medical Technology, Beijing Institute of Technology(北京理工大学医学技术学院);College of Electronics and Information Engineering, Shenzhen University(深圳大学电子与信息工程学院);Department of Computer Science, Faculty of Electrical Engineering and Computer Science, Ningbo University(宁波大学电气工程与计算机科学学院计算机系)

Jiale Chen, Wei Wang, Chongyang Shi, Li Dong, Yuanman Li, Xiping Hu

AI总结:

本文提出基于深度学习的深度鲁棒可逆水印(DRRW),利用整数可逆水印网络和端到端框架解决传统方法设计复杂与鲁棒性差的问题,显著提升了鲁棒性与图像质量,并大幅降低了计算复杂度。

摘要:

鲁棒可逆水印(RRW)在无损信道中能够完美恢复载体图像和水印,同时在有损信道中确保鲁棒的水印提取。现有的RRW方法大多非基于深度学习,面临设计复杂、计算成本高和鲁棒性差的问题,限制了其实际应用。本文提出了深度鲁棒可逆水印(DRRW),一种基于深度学习的RRW方案。DRRW使用整数可逆水印网络(iIWN)可逆地映射整数数据分布,解决了传统RRW的局限性。与需要针对特定失真进行设计的传统RRW不同,DRRW采用编码器-噪声层-解码器框架,通过端到端训练实现自适应鲁棒性。在推理过程中,载体图像和水印映射为溢出的隐写图像和潜变量,通过算术编码压缩为比特流,并通过可逆数据隐藏嵌入以实现无损恢复。我们引入了溢出惩罚损失来减少像素溢出,缩短辅助比特流,同时增强鲁棒性和隐写图像质量。自适应权重调整策略避免了手动水印损失加权,提高了训练稳定性和性能。实验表明,DRRW优于最先进的RRW方法,将鲁棒性提升,并分别将嵌入、提取和恢复复杂度降低了55.14倍、5.95倍和3.57倍。辅助比特流缩小了43.86倍,在16,762张PASCAL VOC 2012图像上成功实现可逆嵌入,推动了实用RRW的发展。DRRW在鲁棒性和质量上超越了不可逆鲁棒水印,同时保持了可逆性。

英文摘要:

Robust Reversible Watermarking (RRW) enables perfect recovery of cover images and watermarks in lossless channels while ensuring robust watermark extraction in lossy channels. Existing RRW methods, mostly non-deep learning-based, face complex designs, high computational costs, and poor robustness, limiting their practical use. This paper proposes Deep Robust Reversible Watermarking (DRRW), a deep learning-based RRW scheme. DRRW uses an Integer Invertible Watermark Network (iIWN) to map integer data distributions invertibly, addressing conventional RRW limitations. Unlike traditional RRW, which needs distortion-specific designs, DRRW employs an encoder-noise layer-decoder framework for adaptive robustness via end-to-end training. In inference, cover image and watermark map to an overflowed stego image and latent variables, compressed by arithmetic coding into a bitstream embedded via reversible data hiding for lossless recovery. We introduce an overflow penalty loss to reduce pixel overflow, shortening the auxiliary bitstream while enhancing robustness and stego image quality. An adaptive weight adjustment strategy avoids manual watermark loss weighting, improving training stability and performance. Experiments show DRRW outperforms state-of-the-art RRW methods, boosting robustness and cutting embedding, extraction, and recovery complexities by 55.14\(\times\), 5.95\(\times\), and 3.57\(\times\), respectively. The auxiliary bitstream shrinks by 43.86\(\times\), with reversible embedding succeeding on 16,762 PASCAL VOC 2012 images, advancing practical RRW. DRRW exceeds irreversible robust watermarking in robustness and quality while maintaining reversibility.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑