On Detecting GANs and Retouching based Synthetic Alterations
01 Oct 2018-pp 1-7
TL;DR: A supervised deep learning algorithm using Convolutional Neural Networks (CNNs) to detect synthetically altered images and yields an accuracy of 99.65% on detecting retouching on the ND-IIITD dataset, which outperforms the previous state of the art.
Abstract: Digitally retouching images has become a popular trend, with people posting altered images on social media and even magazines posting flawless facial images of celebrities. Further, with advancements in Generative Adversarial Networks (GANs), now changing attributes and retouching have become very easy. Such synthetic alterations have adverse effect on face recognition algorithms. While researchers have proposed to detect image tampering, detecting GANs generated images has still not been explored. This paper proposes a supervised deep learning algorithm using Convolutional Neural Networks (CNNs) to detect synthetically altered images. The algorithm yields an accuracy of 99.65% on detecting retouching on the ND-IIITD dataset. It outperforms the previous state of the art which reported an accuracy of 87% on the database. For distinguishing between real images and images generated using GANs, the proposed algorithm yields an accuracy of 99.83%.
Citations
More filters
TL;DR: This survey provides a thorough review of techniques for manipulating face images including DeepFake methods, and methods to detect such manipulations, with special attention to the latest generation of DeepFakes.
502 citations
Cites background or methods from "On Detecting GANs and Retouching ba..."
...proposed in [79] a detection system based on a CNN architecture composed of 6 convolutional layers and 2 fully-connected layers....
[...]
...(2019) [79] Face Patches CNN + SVM Overall Acc....
[...]
Posted Content•
TL;DR: In this paper, the authors provide a thorough review of techniques for manipulating face images including DeepFake methods, and methods to detect such manipulations, including entire face synthesis, identity swap (DeepFakes), attribute manipulation and expression swap.
Abstract: The free access to large-scale public databases, together with the fast progress of deep learning techniques, in particular Generative Adversarial Networks, have led to the generation of very realistic fake content with its corresponding implications towards society in this era of fake news. This survey provides a thorough review of techniques for manipulating face images including DeepFake methods, and methods to detect such manipulations. In particular, four types of facial manipulation are reviewed: i) entire face synthesis, ii) identity swap (DeepFakes), iii) attribute manipulation, and iv) expression swap. For each manipulation group, we provide details regarding manipulation techniques, existing public databases, and key benchmarks for technology evaluation of fake detection methods, including a summary of results from those evaluations. Among all the aspects discussed in the survey, we pay special attention to the latest generation of DeepFakes, highlighting its improvements and challenges for fake detection.
In addition to the survey information, we also discuss open issues and future trends that should be considered to advance in the field.
42 citations
TL;DR: A retouching detection system based on the analysis of photo response non-uniformity (PRNU) that is shown to robustly distinguish between bona fide and retouched images achieving an average detection equal error rate of 13.7% across all retoucheding algorithms.
Abstract: Nowadays, many facial images are acquired using smart phones. To ensure the best outcome, users frequently retouch these images before sharing them, e.g. via social media. Modifications resulting from used retouching algorithms might be a challenge for face recognition technologies. Towards deploying robust face recognition as well as enforcing anti-photoshop legislations, a reliable detection of retouched face images is needed. In this work, the effects of facial retouching on face recognition are investigated. A qualitative assessment of 32 beautification apps is conducted. Based on this assessment five apps are chosen which are used to create a database of 800 beautified face images. Biometric performance is measured before and after retouching using a commercial face recognition system. Subsequently, a retouching detection system based on the analysis of photo response non-uniformity (PRNU) is presented. Specifically, scores obtained from analysing spatial and spectral features extracted from PRNU patterns across image cells are fused. In a scenario, in which unaltered bona fide images are compressed to the average sizes of the retouched images using JPEG, the proposed PRNU-based detection scheme is shown to robustly distinguish between bona fide and retouched images achieving an average detection equal error rate of 13.7% across all retouching algorithms.
36 citations
Cites background or methods from "On Detecting GANs and Retouching ba..."
...[5] reported a significant decrease in the detection accuracy in case image compression is only applied to retouched images during training....
[...]
...To the best of the authors’ knowledge there exists only one publicly available database1 of retouched face images which has been used in [3, 5]....
[...]
...[5] ND-IIITD Retouched Faces CNN with SVM – – 99....
[...]
Posted Content•
TL;DR: In this article, a comprehensive review and detailed analysis of existing tools and machine learning (ML) based approaches for deepfake generation and the methodologies used to detect such manipulations for the detection and generation of both audio and video deepfakes.
Abstract: Easy access to audio-visual content on social media, combined with the availability of modern tools such as Tensorflow or Keras, open-source trained models, and economical computing infrastructure, and the rapid evolution of deep-learning (DL) methods, especially Generative Adversarial Networks (GAN), have made it possible to generate deepfakes to disseminate disinformation, revenge porn, financial frauds, hoaxes, and to disrupt government functioning. The existing surveys have mainly focused on deepfake video detection only. No attempt has been made to review approaches for detection and generation of both audio and video deepfakes. This paper provides a comprehensive review and detailed analysis of existing tools and machine learning (ML) based approaches for deepfake generation and the methodologies used to detect such manipulations for the detection and generation of both audio and video deepfakes. For each category of deepfake, we discuss information related to manipulation approaches, current public datasets, and key standards for the performance evaluation of deepfake detection techniques along with their results. Additionally, we also discuss open challenges and enumerate future directions to guide future researchers on issues that need to be considered to improve the domains of both the deepfake generation and detection. This work is expected to assist the readers in understanding the creation and detection mechanisms of deepfake, along with their current limitations and future direction.
29 citations
Posted Content•
TL;DR: In this paper, a combination of co-occurrence matrices and deep learning is proposed to detect GAN generated fake images using three color channels in the pixel domain and train a model using a deep convolutional neural network (CNN) framework.
Abstract: The advent of Generative Adversarial Networks (GANs) has brought about completely novel ways of transforming and manipulating pixels in digital images. GAN based techniques such as Image-to-Image translations, DeepFakes, and other automated methods have become increasingly popular in creating fake images. In this paper, we propose a novel approach to detect GAN generated fake images using a combination of co-occurrence matrices and deep learning. We extract co-occurrence matrices on three color channels in the pixel domain and train a model using a deep convolutional neural network (CNN) framework. Experimental results on two diverse and challenging GAN datasets comprising more than 56,000 images based on unpaired image-to-image translations (cycleGAN [1]) and facial attributes/expressions (StarGAN [2]) show that our approach is promising and achieves more than 99% classification accuracy in both datasets. Further, our approach also generalizes well and achieves good results when trained on one dataset and tested on the other.
25 citations
References
More filters
01 Oct 2017
TL;DR: CycleGAN as discussed by the authors learns a mapping G : X → Y such that the distribution of images from G(X) is indistinguishable from the distribution Y using an adversarial loss.
Abstract: Image-to-image translation is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image using a training set of aligned image pairs. However, for many tasks, paired training data will not be available. We present an approach for learning to translate an image from a source domain X to a target domain Y in the absence of paired examples. Our goal is to learn a mapping G : X → Y such that the distribution of images from G(X) is indistinguishable from the distribution Y using an adversarial loss. Because this mapping is highly under-constrained, we couple it with an inverse mapping F : Y → X and introduce a cycle consistency loss to push F(G(X)) ≈ X (and vice versa). Qualitative results are presented on several tasks where paired training data does not exist, including collection style transfer, object transfiguration, season transfer, photo enhancement, etc. Quantitative comparisons against several prior methods demonstrate the superiority of our approach.
11,682 citations
Posted Content•
TL;DR: Conditional Adversarial Network (CA) as discussed by the authors is a general-purpose solution to image-to-image translation problems, which can be used to synthesize photos from label maps, reconstructing objects from edge maps, and colorizing images, among other tasks.
Abstract: We investigate conditional adversarial networks as a general-purpose solution to image-to-image translation problems. These networks not only learn the mapping from input image to output image, but also learn a loss function to train this mapping. This makes it possible to apply the same generic approach to problems that traditionally would require very different loss formulations. We demonstrate that this approach is effective at synthesizing photos from label maps, reconstructing objects from edge maps, and colorizing images, among other tasks. Indeed, since the release of the pix2pix software associated with this paper, a large number of internet users (many of them artists) have posted their own experiments with our system, further demonstrating its wide applicability and ease of adoption without the need for parameter tweaking. As a community, we no longer hand-engineer our mapping functions, and this work suggests we can achieve reasonable results without hand-engineering our loss functions either.
11,127 citations
Proceedings Article•
31 Mar 2010TL;DR: The objective here is to understand better why standard gradient descent from random initialization is doing so poorly with deep neural networks, to better understand these recent relative successes and help design better algorithms in the future.
Abstract: Whereas before 2006 it appears that deep multilayer neural networks were not successfully trained, since then several algorithms have been shown to successfully train them, with experimental results showing the superiority of deeper vs less deep architectures. All these experimental results were obtained with new initialization or training mechanisms. Our objective here is to understand better why standard gradient descent from random initialization is doing so poorly with deep neural networks, to better understand these recent relative successes and help design better algorithms in the future. We first observe the influence of the non-linear activations functions. We find that the logistic sigmoid activation is unsuited for deep networks with random initialization because of its mean value, which can drive especially the top hidden layer into saturation. Surprisingly, we find that saturated units can move out of saturation by themselves, albeit slowly, and explaining the plateaus sometimes seen when training neural networks. We find that a new non-linearity that saturates less can often be beneficial. Finally, we study how activations and gradients vary across layers and during training, with the idea that training may be more difficult when the singular values of the Jacobian associated with each layer are far from 1. Based on these considerations, we propose a new initialization scheme that brings substantially faster convergence. 1 Deep Neural Networks Deep learning methods aim at learning feature hierarchies with features from higher levels of the hierarchy formed by the composition of lower level features. They include Appearing in Proceedings of the 13 International Conference on Artificial Intelligence and Statistics (AISTATS) 2010, Chia Laguna Resort, Sardinia, Italy. Volume 9 of JMLR: WC Weston et al., 2008). Much attention has recently been devoted to them (see (Bengio, 2009) for a review), because of their theoretical appeal, inspiration from biology and human cognition, and because of empirical success in vision (Ranzato et al., 2007; Larochelle et al., 2007; Vincent et al., 2008) and natural language processing (NLP) (Collobert & Weston, 2008; Mnih & Hinton, 2009). Theoretical results reviewed and discussed by Bengio (2009), suggest that in order to learn the kind of complicated functions that can represent high-level abstractions (e.g. in vision, language, and other AI-level tasks), one may need deep architectures. Most of the recent experimental results with deep architecture are obtained with models that can be turned into deep supervised neural networks, but with initialization or training schemes different from the classical feedforward neural networks (Rumelhart et al., 1986). Why are these new algorithms working so much better than the standard random initialization and gradient-based optimization of a supervised training criterion? Part of the answer may be found in recent analyses of the effect of unsupervised pretraining (Erhan et al., 2009), showing that it acts as a regularizer that initializes the parameters in a “better” basin of attraction of the optimization procedure, corresponding to an apparent local minimum associated with better generalization. But earlier work (Bengio et al., 2007) had shown that even a purely supervised but greedy layer-wise procedure would give better results. So here instead of focusing on what unsupervised pre-training or semi-supervised criteria bring to deep architectures, we focus on analyzing what may be going wrong with good old (but deep) multilayer neural networks. Our analysis is driven by investigative experiments to monitor activations (watching for saturation of hidden units) and gradients, across layers and across training iterations. We also evaluate the effects on these of choices of activation function (with the idea that it might affect saturation) and initialization procedure (since unsupervised pretraining is a particular form of initialization and it has a drastic impact).
9,500 citations
"On Detecting GANs and Retouching ba..." refers methods in this paper
...The model weights are initialized using Xavier’s method as shown by Glorot and Bengio [10]....
[...]
Proceedings Article•
23 Feb 2016TL;DR: In this article, the authors show that training with residual connections accelerates the training of Inception networks significantly, and they also present several new streamlined architectures for both residual and non-residual Inception Networks.
Abstract: Very deep convolutional networks have been central to the largest advances in image recognition performance in recent years. One example is the Inception architecture that has been shown to achieve very good performance at relatively low computational cost. Recently, the introduction of residual connections in conjunction with a more traditional architecture has yielded state-of-the-art performance in the 2015 ILSVRC challenge; its performance was similar to the latest generation Inception-v3 network. This raises the question: Are there any benefits to combining Inception architectures with residual connections? Here we give clear empirical evidence that training with residual connections accelerates the training of Inception networks significantly. There is also some evidence of residual Inception networks outperforming similarly expensive Inception networks without residual connections by a thin margin. We also present several new streamlined architectures for both residual and non-residual Inception networks. These variations improve the single-frame recognition performance on the ILSVRC 2012 classification task significantly. We further demonstrate how proper activation scaling stabilizes the training of very wide residual Inception networks. With an ensemble of three residual and one Inception-v4 networks, we achieve 3.08% top-5 error on the test set of the ImageNet classification (CLS) challenge.
4,051 citations
Posted Content•
TL;DR: Wide residual networks (WRNs) as mentioned in this paper decrease the depth and increase the width of residual networks, which achieves state-of-the-art results on CIFAR, SVHN, and ImageNet.
Abstract: Deep residual networks were shown to be able to scale up to thousands of layers and still have improving performance. However, each fraction of a percent of improved accuracy costs nearly doubling the number of layers, and so training very deep residual networks has a problem of diminishing feature reuse, which makes these networks very slow to train. To tackle these problems, in this paper we conduct a detailed experimental study on the architecture of ResNet blocks, based on which we propose a novel architecture where we decrease depth and increase width of residual networks. We call the resulting network structures wide residual networks (WRNs) and show that these are far superior over their commonly used thin and very deep counterparts. For example, we demonstrate that even a simple 16-layer-deep wide residual network outperforms in accuracy and efficiency all previous deep residual networks, including thousand-layer-deep networks, achieving new state-of-the-art results on CIFAR, SVHN, COCO, and significant improvements on ImageNet. Our code and models are available at this https URL
3,531 citations