# CycleGAN, a Master of Steganography

arXiv: https://arxiv.org/abs/1712.02950 · HTML: https://mlsanity.com/paper/1712.02950/

## Key facts

- Cited by 65 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Dec 2017.

## Built on by (most cited first, 30 of 65)

- [The Alignment Problem from a Deep Learning Perspective](https://mlsanity.com/paper/2209.00626/) (2022)
- [Variational Mixture-of-Experts Autoencoders for Multi-Modal Deep Generative Models](https://mlsanity.com/paper/1911.03393/) (2019)
- [Learning to generate line drawings that convey geometry and semantics](https://mlsanity.com/paper/2203.12691/) (2022)
- [Preventing Language Models From Hiding Their Reasoning](https://mlsanity.com/paper/2310.18512/) (2023)
- [LEO: Generative Latent Image Animator for Human Video Synthesis](https://mlsanity.com/paper/2305.03989/) (2023)
- [High-Fidelity and Arbitrary Face Editing](https://mlsanity.com/paper/2103.15814/) (2021)
- [DeFlow: Learning Complex Image Degradations from Unpaired Data with Conditional Flows](https://mlsanity.com/paper/2101.05796/) (2021)
- [Controllable Generation from Pre-trained Language Models via Inverse Prompting](https://mlsanity.com/paper/2103.10685/) (2021)
- [Segmentation-Renormalized Deep Feature Modulation for Unpaired Image Harmonization](https://mlsanity.com/paper/2102.06315/) (2021)
- [FUN-SIS: a Fully UNsupervised approach for Surgical Instrument Segmentation](https://mlsanity.com/paper/2202.08141/) (2022)
- [Learn from Unpaired Data for Image Restoration: A Variational Bayes Approach](https://mlsanity.com/paper/2204.10090/) (2022)
- [Synthetic ID Card Image Generation for Improving Presentation Attack Detection](https://mlsanity.com/paper/2211.00098/) (2022)
- [StegoGAN: Leveraging Steganography for Non-Bijective Image-to-Image Translation](https://mlsanity.com/paper/2403.20142/) (2024)
- [Generative Steganography Network](https://mlsanity.com/paper/2207.13867/) (2022)
- [Dataset Condensation via Generative Model](https://mlsanity.com/paper/2309.07698/) (2023)
- [Detection of AI Deepfake and Fraud in Online Payments Using GAN-Based Models](https://mlsanity.com/paper/2501.07033/) (2025)
- [Inverse Rendering of Translucent Objects using Physical and Neural Renderers](https://mlsanity.com/paper/2305.08336/) (2023)
- [Exploring Semantic Consistency in Unpaired Image Translation to Generate Data for Surgical Applications](https://mlsanity.com/paper/2309.03048/) (2023)
- [Towards Better Adversarial Synthesis of Human Images from Text](https://mlsanity.com/paper/2107.01869/) (2021)
- [CYCLE-INSTRUCT: Fully Seed-Free Instruction Tuning via Dual Self-Training and Cycle Consistency](https://mlsanity.com/paper/2508.16100/) (2025)
- [A Chain-of-thought Reasoning Breast Ultrasound Dataset Covering All Histopathology Categories](https://mlsanity.com/paper/2509.17046/) (2025)
- [Generating Synthetic Handwritten Historical Documents With OCR Constrained GANs](https://mlsanity.com/paper/2103.08236/) (2021)
- [Quality Metric Guided Portrait Line Drawing Generation from Unpaired Training Data](https://mlsanity.com/paper/2202.03678/) (2022)
- [Self adversarial attack as an augmentation method for immunohistochemical stainings](https://mlsanity.com/paper/2103.11362/) (2021)
- [Unsupervised Detection of Cancerous Regions in Histology Imagery using Image-to-Image Translation](https://mlsanity.com/paper/2104.13786/) (2021)
- [CLTS-GAN: Color-Lighting-Texture-Specular Reflection Augmentation for Colonoscopy](https://mlsanity.com/paper/2206.14951/) (2022)
- [A Pattern Language for Machine Learning Tasks](https://mlsanity.com/paper/2407.02424/) (2024)
- [VaseVQA-3D: Benchmarking 3D VLMs on Ancient Greek Pottery](https://mlsanity.com/paper/2510.04479/) (2025)
- [Style Agnostic 3D Reconstruction via Adversarial Style Transfer](https://mlsanity.com/paper/2110.10784/) (2021)
- [Policy Gradient Stock GAN for Realistic Discrete Order Data Generation in Financial Markets](https://mlsanity.com/paper/2204.13338/) (2022)

## Abstract (arXiv)

CycleGAN (Zhu et al. 2017) is one recent successful approach to learn a transformation between two image distributions. In a series of experiments, we demonstrate an intriguing property of the model: CycleGAN learns to "hide" information about a source image into the images it generates in a nearly imperceptible, high-frequency signal. This trick ensures that the generator can recover the original sample and thus satisfy the cyclic consistency requirement, while the generated image remains realistic. We connect this phenomenon with adversarial attacks by viewing CycleGAN's training procedure as training a generator of adversarial examples and demonstrate that the cyclic consistency loss causes CycleGAN to be especially vulnerable to adversarial attacks.