# Learning Generalizable and Efficient Image Watermarking via Hierarchical Two-Stage Optimization

arXiv: https://arxiv.org/abs/2508.08667 · HTML: https://mlsanity.com/paper/2508.08667/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 61 references that are in the map.
- First posted on arXiv 12 Aug 2025.

## Builds on (most cited first, 25 of 25)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation](https://mlsanity.com/paper/1806.10293/) (2018)
- [A Watermark for Large Language Models](https://mlsanity.com/paper/2301.10226/) (2023)
- [HiDDeN: Hiding Data With Deep Networks](https://mlsanity.com/paper/1807.09937/) (2018)
- [GenImage: A Million-Scale Benchmark for Detecting AI-Generated Image](https://mlsanity.com/paper/2306.08571/) (2023)
- [The Stable Signature: Rooting Watermarks in Latent Diffusion Models](https://mlsanity.com/paper/2303.15435/) (2023)
- [Artificial Fingerprinting for Generative Models: Rooting Deepfake Attribution in Training Data](https://mlsanity.com/paper/2007.08457/) (2020)
- [Gaussian Shading: Provable Performance-Lossless Image Watermarking for Diffusion Models](https://mlsanity.com/paper/2404.04956/) (2024)
- [Unbiased Watermark for Large Language Models](https://mlsanity.com/paper/2310.10669/) (2023)
- [MBRS : Enhancing Robustness of DNN-based Watermarking by Mini-Batch of Real and Simulated JPEG Compression](https://mlsanity.com/paper/2108.08211/) (2021)
- [Watermarking Images in Self-Supervised Latent Spaces](https://mlsanity.com/paper/2112.09581/) (2021)
- [Human-Art: A Versatile Human-Centric Dataset Bridging Natural and Artificial Scenes](https://mlsanity.com/paper/2303.02760/) (2023)
- [Model Watermarking for Image Processing Networks](https://mlsanity.com/paper/2002.11088/) (2020)
- [Protecting Intellectual Property of Generative Adversarial Networks from Ambiguity Attack](https://mlsanity.com/paper/2102.04362/) (2021)
- [ROBIN: Robust and Invisible Watermarks for Diffusion Models with Adversarial Optimization](https://mlsanity.com/paper/2411.03862/) (2024)
- [Supervised GAN Watermarking for Intellectual Property Protection](https://mlsanity.com/paper/2209.03466/) (2022)
- [Latent Watermark: Inject and Detect Watermarks in Latent Diffusion Space](https://mlsanity.com/paper/2404.00230/) (2024)
- [Free Fine-tuning: A Plug-and-Play Watermarking Scheme for Deep Neural Networks](https://mlsanity.com/paper/2210.07809/) (2022)
- [Practical Deep Dispersed Watermarking with Synchronization and Fusion](https://mlsanity.com/paper/2310.14532/) (2023)
- [De-END: Decoder-driven Watermarking Network](https://mlsanity.com/paper/2206.13032/) (2022)
- [SSyncOA: Self-synchronizing Object-aligned Watermarking to Resist Cropping-paste Attacks](https://mlsanity.com/paper/2405.03458/) (2024)
- [Achieving Resolution-Agnostic DNN-based Image Watermarking: A Novel Perspective of Implicit Neural Representation](https://mlsanity.com/paper/2405.08340/) (2024)

## Abstract (arXiv)

Deep image watermarking, which refers to enabling imperceptible watermark embedding and reliable extraction in cover images, has been shown to be effective for copyright protection of image assets. However, existing methods face limitations in simultaneously satisfying three essential criteria for generalizable watermarking: (1) invisibility (imperceptible hiding of watermarks), (2) robustness (reliable watermark recovery under diverse conditions), and (3) broad applicability (low latency in the watermarking process). To address these limitations, we propose a Hierarchical Watermark Learning (HiWL) framework, a two-stage optimization that enables a watermarking model to simultaneously achieve all three criteria. In the first stage, distribution alignment learning is designed to establish a common latent space with two constraints: (1) visual consistency between watermarked and non-watermarked images, and (2) information invariance across watermark latent representations. In this way, multimodal inputs -- including watermark messages (binary codes) and cover images (RGB pixels) -- can be effectively represented, ensuring both the invisibility of watermarks and robustness in the watermarking process. In the second stage, we employ generalized watermark representation learning to separate a unique representation of the watermark from the marked image in RGB space. Once trained, the HiWL model effectively learns generalizable watermark representations while maintaining broad applicability. Extensive experiments demonstrate the effectiveness of the proposed method. Specifically, it achieves 7.6% higher accuracy in watermark extraction compared to existing methods, while maintaining extremely low latency (processing 1000 images in 1 second).