「CV」 数字水印资源汇总

在不影响数字载体（包括多媒体——图片、视频、音频，文档，软件等）使用价值的情况下，将标识信息嵌入到载体中；
数字水印侧重于水印信息的鲁棒性，隐写侧重中水印信息的隐蔽性；

图1：数字水印流程

Digital Watermarking · Steganography

1 数字水印

1.1 综述

1. Watermarking of Digital Images
   2012-06-03 paper
   $\bullet \bullet$
   关于传统算法的水印综述，内容非常全面广泛，除了没有覆盖到 DL；
   貌似是 99 年的论文，因为原稿丢了，12 年重新出版了这个；

2. A Survey of Digital Watermarking Techniques and its Applications
   2014-07-17 paper
   $\bullet \bullet$ 入门
   关于传统水印算法的总数，总共 4 页，真是短小精悍；看完会对水印有个整体认识的，非常推荐作为水印入门读物；

   可惜了我花了两天时间要死要活地折腾出一个Survey，要是早看到这篇，分分钟搞定，看来还是综述文章最容易入门；

3. Analysis of Visible and Invisible Image Watermarking – A Review
   2016-08 paper
   针对传统方法的综述，主讲了 DWT、DCT 和 DFT；

   文章内容较少，不是一篇优质的 review，应该是频域水印算法回顾；

4. A Comprehensive Survey of Watermarking Relational Databases Research
   2018-01-25 paper
   数据库和数字指纹综述；

   似乎已经超出 CV 领域，不过也有借鉴意义，空了可以看；

1.2 理论

1. Steganography Security: Principle and Practice
   2018-06-10 paper
   对比了隐写术和数字水印的安全性差异；

1.3 传统方法

1. A New Method For Digital Watermarking Based on Combination of DCT and PCA
   TELFOR 2014 2015-09-10 paper

2. Digital image watermarking using normal matrices
   2015-05-23 paper

3. OR-Benchmark: An Open and Reconfigurable Digital Watermarking Benchmarking Framework
   2015-05-31 paper

1.4 深度学习

1.4.1 通用

1. Watermarking Deep Neural Networks for Embedded Systems
   ICCAD 2018 2018 paper

2. Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooring
   2018-02-13 paper

3. Protecting Intellectual Property of Deep Neural Networks with Watermarking
   AsiaCCS 2018 paper

4. On the Robustness of the Backdoor-based Watermarking in Deep Neural Networks
   2019-06-18 paper

1.4.2 图像

1.4.3 音频

1.4.4 视频

1.4.5 网络模型

给模型打水印；

1. DeepSigns: An End-to-End Watermarking Framework for Protecting the Ownership of Deep Neural Networks
   paper

2. Embedding Watermarks into Deep Neural Networks
   ICMR 2017 2017-01-15 paper | keras

3. Digital Watermarking for Deep Neural Networks
   2018-02-06 paper

4. Have You Stolen My Model? Evasion Attacks Against Deep Neural Network Watermarking Techniques
   2018-09-03 paper

5. Cryptographic key distribution over a public network via variance-based watermarking in compressive measurements
   2019-03-30 paper

6. Effectiveness of Distillation Attack and Countermeasure on Neural Network Watermarking
   2019-06-14 paper

1.4.6 跨模态

2 隐写术

1.1 综述

1. Deep Learning in steganography and steganalysis from 2015 to 2018
   2019-03-31 paper

2. Recent Advances of Image Steganography with Generative Adversarial Networks
   2019-06-18 paper
   GAN 在隐写术里的应用；

2.2 理论

1. Steganography Security: Principle and Practice
   2018-06-10 paper
   对比了隐写术和数字水印的安全性差异；

2.3 传统方法

2.4 深度学习

2.4.1 通用

1. Moving Steganography and Steganalysis from the Laboratory into the Real World
   2013-06-20 paper
   显示应用；

2. SSGAN: Secure Steganography Based on Generative Adversarial Networks
   2017-06-06 paper

3. Convolutional Neural Network Steganalysis’s Application to Steganography
   2017-11-06 paper

4. Generative Steganography with Kerckhoffs’ Principle based on Generative Adversarial Networks
   2017-11-14 paper

5. CycleGAN, a Master of Steganography
   NIPS 2017 2017-12-08 paper

6. SteganoGAN: High Capacity Image Steganography with GANs
   2019-01-12 paper | pytorch-offical | 文档

7. Learning Symmetric and Asymmetric Steganography via Adversarial Training
   2019-03-13 paper

8. BASN – Learning Steganography with Binary Attention Mechanism
   2019-07-09 paper

9. Steganography using a 3 player game
   2019-07-14 paper

2.4.2 图像

1. Generating steganographic images via adversarial training
   2017-03-01 paper

2. End-to-end Trained CNN Encode-Decoder Networks for Image Steganography
   2017-11-20 paper

3. A Novel Convolutional Neural Network for Image Steganalysis with Shared Normalization
   2017-11-20 paper

4. Deep Learning Hierarchical Representations for Image Steganalysis
   2017-11 paper | pytorch

5. CNN Based Adversarial Embedding with Minimum Alteration for Image Steganography
   IEEE 2018-03-24 paper

6. Spatial Image Steganography Based on Generative Adversarial Network
   2018-04-21 paper

7. StegNet: Mega Image Steganography Capacity with Deep Convolutional Network
   2018-06-17 paper | tensorflow-offical

8. Invisible Steganography via Generative Adversarial Networks
   2018-07-23 paper

9. Efficient feature learning and multi-size image steganalysis based on CNN
   2018-07-30 paper
   ZhuNet，任意尺寸输入，Inception 有效性证明；

10. StegoAppDB: a Steganography Apps Forensics Image Database
    2019-04-19 paper
    数据集；

11. EncryptGAN: Image Steganography with Domain Transform
    2019-05-28 paper

2.4.3 音频

1. Steganography between Silence Intervals of Audio in Video Content Using Chaotic Maps
   2016-10-14 paper
   音轨中静默片段放水印；

2. StegIbiza: Steganography in Club Music Implemented in Python
   2017-05-22 paper

3. AAG-Stega: Automatic Audio Generation-based Steganography
   AAAi 2019 2018-09-10 paper

4. Hide and Speak: Deep Neural Networks for Speech Steganography
   2019-02-07 paper

5. Heard More Than Heard: An Audio Steganography Method Based on GAN
   2019-07-11 paper

2.4.4 视频

1. Convolutional Video Steganography with Temporal Residual Modeling
   2018-06-08 paper

2.4.5 网络模型

给模型打水印；

1. DeepStego: Protecting Intellectual Property of Deep Neural Networks by Steganography
   2019-03-05 paper

2.4.6 跨模态

1. Provably Secure Steganography on Generative Media
   2018-11-09 paper

3 攻击

1. Fraternal Twins: Unifying Attacks on Machine Learning and Digital Watermarking
   2017-03-16 paper

2. Attacks on Digital Watermarks for Deep Neural Networks
   2019 paper

4 防篡改

video tampering

1. Computer-Aided Annotation for Video Tampering Dataset of Forensic Research
   2018-02-07 paper

2. ARCHANGEL: Tamper-proofing Video Archives using Temporal Content Hashes on the Blockchain
   CVPR 2019 workshop 2019-04-26 paper

---

TOP

附录

A 参考资料

B 数据集

数据集	训练集（万张）	验证集（万张）	测试集（万张）	分辨率
BOSS	0.9	0.1		512×512