• Corpus ID: 235266180

Markpainting: Adversarial Machine Learning meets Inpainting

@inproceedings{Khachaturov2021MarkpaintingAM,
  title={Markpainting: Adversarial Machine Learning meets Inpainting},
  author={David Khachaturov and Ilia Shumailov and Yiren Zhao and Nicolas Papernot and Ross Anderson},
  booktitle={ICML},
  year={2021}
}
Inpainting is a learned interpolation technique that is based on generative modeling and used to populate masked or missing pieces in an image; it has wide applications in picture editing and retouching. Recently, inpainting started being used for watermark removal, raising concerns. In this paper we study how to manipulate it using our markpainting technique. First, we show how an image owner with access to an inpainting model can augment their image in such a way that any attempt to edit it… 
[PDF] Semantic Reader
8 Citations
Background Citations
6
Methods Citations
3

8 Citations

NoiLIn: Improving Adversarial Training and Correcting Stereotype of Noisy Labels
TLDR
A simple but effective method—NoiLIn that randomly injects NLs into training data at each training epoch and dynamically increases the NL injection rate once robust overfitting occurs can mitigate the AT’s undesirable issue of robust over-tting and improve the generalization of the state-of-the-art AT methods.
Hiding Images into Images with Real-world Robustness
TLDR
A generative deep network based method for hiding images into images while assuring high-quality extraction from the destructive synthesized images, which is the first to robustly hide three secret images.
RWN: Robust Watermarking Network for Image Cropping Localization
TLDR
A novel robust watermarking network (RWN) for image crop localization using an anti-crop processor that embeds a watermark into a target image and extends the scheme to detect tampering attack on the attacked image.
No way to crop: On robust image crop localization
TLDR
This paper trains an anti-crop processor that embeds a watermark into a target image that is then posted on the social network instead of the original image, and demonstrates that this scheme is the first to provide high-accuracy and robust image crop localization.
Robust Watermarking for Video Forgery Detection with Improved Imperceptibility and Robustness
TLDR
A video watermarking network for tampering localization is proposed and it is proposed to mimic video compression by ensembling simulation results of other typical attacks, e.g., JPEG compression and blurring, as an approximation.
Leveraging Adversarial Examples to Quantify Membership Information Leakage
TLDR
This work develops a novel approach to address the problem of membership inference in pattern recognition models, relying on information provided by adversarial examples, by measuring the magnitude of a perturbation necessary to build an adversarial example.
Are Deepfakes Concerning? Analyzing Conversations of Deepfakes on Reddit and Exploring Societal Implications
TLDR
It is proposed that there are implications for Human Computer Interaction (HCI) to mitigate the harm created from deepfakes and build a community that supports creating and sharing deepfake artifacts and building a marketplace regardless of the consequences.
Robust Image Protection Countering Cropping Manipulation
TLDR
Experiments demonstrate that CLR-Net can accurately localize the cropping as well as recover the details of the cropped-out regions with both high quality and reputation, despite of the presence of image processing attacks of varied types.

References

SHOWING 1-10 OF 45 REFERENCES
EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning
TLDR
A new approach for image inpainting that does a better job of reproducing filled regions exhibiting fine details is developed and outperforms current state-of-the-art techniques quantitatively and qualitatively.
Image inpainting
TLDR
A novel algorithm for digital inpainting of still images that attempts to replicate the basic techniques used by professional restorators, and does not require the user to specify where the novel information comes from.
Generative Image Inpainting with Contextual Attention
TLDR
This work proposes a new deep generative model-based approach which can not only synthesize novel image structures but also explicitly utilize surrounding image features as references during network training to make better predictions.
Image Inpainting with Contextual Reconstruction Loss
TLDR
Experimental results demonstrate that the proposed inpainting model with the CR loss compares favourably against the state-of-the-arts in terms of quantitative and visual performance.
Context Encoders: Feature Learning by Inpainting
TLDR
It is found that a context encoder learns a representation that captures not just appearance but also the semantics of visual structures, and can be used for semantic inpainting tasks, either stand-alone or as initialization for non-parametric methods.
Recurrent Feature Reasoning for Image Inpainting
TLDR
A Recurrent Feature Reasoning (RFR) network which is mainly constructed by a plug-and-play Recurrent feature Reasoning module and a Knowledge Consistent Attention (KCA) module, which recurrently infers the hole boundaries of the convolutional feature maps and uses them as clues for further inference.
ZOO: Zeroth Order Optimization Based Black-box Attacks to Deep Neural Networks without Training Substitute Models
TLDR
An effective black-box attack that also only has access to the input (images) and the output (confidence scores) of a targeted DNN is proposed, sparing the need for training substitute models and avoiding the loss in attack transferability.
Fast Face-Swap Using Convolutional Neural Networks
TLDR
A new loss function is devised that enables the network to produce highly photorealistic results by making face swap work in real-time with no input from the user.
Foreground-Aware Image Inpainting
  • Wei Xiong, Jiahui Yu, Jiebo Luo
  • Computer Science
    2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2019
TLDR
This work proposes a foreground-aware image inpainting system that explicitly disentangles structure inference and content completion, and shows that by such disentanglement, the contour completion model predicts reasonable contours of objects, and further substantially improves the performance of image inPainting.
Learning to Incorporate Structure Knowledge for Image Inpainting
TLDR
This paper develops a multi-task learning framework that attempts to incorporate the image structure knowledge to assist image inpainting and takes advantage of the structure knowledge and outperforms several state-of-the-art methods on benchmark datasets quantitatively and qualitatively.
...
...