Search

Detecting AI-generated images has become an extraordinarily difficult challenge as new generative architectures emerge on a daily basis with more and more capabilities and unprecedented realism. New versions of many commercial tools, such as DALLE, Midjourney, and Stable Diffusion, have been released recently, and it is impractical to continually update and retrain supervised forensic detectors to handle such a large variety of models. To address this challenge, we propose a zero-shot entropy-based detector (ZED) that neither needs AI-generated training data nor relies on knowledge of generative architectures to artificially synthesize their artifacts. Inspired by recent works on machine-generated text detection, our idea is to measure how surprising the image under analysis is compared to a model of real images. To this end, we rely on a lossless image encoder that estimates the probability distribution of each pixel given its context. To ensure computational efficiency, the encoder has a multi-resolution architecture and contexts comprise mostly pixels of the lower-resolution version of the image.Since only real images are needed to learn the model, the detector is independent of generator architectures and synthetic training data. Using a single discriminative feature, the proposed detector achieves state-of-the-art performance. On a wide variety of generative models it achieves an average improvement of more than 3% over the SoTA in terms of accuracy. Code is available at https://grip-unina.github.io/ZED/.

In recent years, many forensic detectors have been proposed to detect AI-generated images and prevent their use for malicious purposes. Convolutional neural networks (CNNs) have long been the dominant architecture in this field and have been the subject of intense study. However, recently proposed Transformer-based detectors have been shown to match or even outperform CNN-based detectors, especially in terms of generalization. In this paper, we study the adversarial robustness of AI-generated image detectors, focusing on Contrastive Language-Image Pretraining (CLIP)-based methods that rely on Visual Transformer (ViT) backbones and comparing their performance with CNN-based methods. We study the robustness to different adversarial attacks under a variety of conditions and analyze both numerical results and frequency-domain patterns. CLIP-based detectors are found to be vulnerable to white-box attacks just like CNN-based detectors. However, attacks do not easily transfer between CNN-based and CLIP-based methods. This is also confirmed by the different distribution of the adversarial noise patterns in the frequency domain. Overall, this analysis provides new insights into the properties of forensic detectors that can help to develop more effective strategies.

Hyperspectral pansharpening consists of fusing a high-resolution panchromatic band and a low-resolution hyperspectral image to obtain a new image with high resolution in both the spatial and spectral domains. These remote sensing products are valuable for a wide range of applications, driving ever growing research efforts. Nonetheless, results still do not meet application demands. In part, this comes from the technical complexity of the task: compared to multispectral pansharpening, many more bands are involved, in a spectral range only partially covered by the panchromatic component and with overwhelming noise. However, another major limiting factor is the absence of a comprehensive framework for the rapid development and accurate evaluation of new methods. This paper attempts to address this issue. We started by designing a dataset large and diverse enough to allow reliable training (for data-driven methods) and testing of new methods. Then, we selected a set of state-of-the-art methods, following different approaches, characterized by promising performance, and reimplemented them in a single PyTorch framework. Finally, we carried out a critical comparative analysis of all methods, using the most accredited quality indicators. The analysis highlights the main limitations of current solutions in terms of spectral/spatial quality and computational efficiency, and suggests promising research directions. To ensure full reproducibility of the results and support future research, the framework (including codes, evaluation procedures and links to the dataset) is shared on https://github.com/matciotola/hyperspectral_pansharpening_toolbox, as a single Python-based reference benchmark toolbox.

Generalization is a main issue for current audio deepfake detectors, which struggle to provide reliable results on out-of-distribution data. Given the speed at which more and more accurate synthesis methods are developed, it is very important to design techniques that work well also on data they were not trained for. In this paper we study the potential of large-scale pre-trained models for audio deepfake detection, with special focus on generalization ability. To this end, the detection problem is reformulated in a speaker verification framework and fake audios are exposed by the mismatch between the voice sample under test and the voice of the claimed identity. With this paradigm, no fake speech sample is necessary in training, cutting off any link with the generation method at the root, and ensuring full generalization ability. Features are extracted by general-purpose large pre-trained models, with no need for training or fine-tuning on specific fake detection or speaker verification datasets. At detection time only a limited set of voice fragments of the identity under test is required. Experiments on several datasets widespread in the community show that detectors based on pre-trained models achieve excellent performance and show strong generalization ability, rivaling supervised methods on in-distribution data and largely overcoming them on out-of-distribution data.

In this work we present an overview of approaches for the detection and attribution of synthetic images and highlight their strengths and weaknesses. We also point out and discuss hot topics in this field and outline promising directions for future research.

The aim of this work is to explore the potential of pre-trained vision-language models (VLMs) for universal detection of AI-generated images. We develop a lightweight detection strategy based on CLIP features and study its performance in a wide variety of challenging scenarios. We find that, contrary to previous beliefs, it is neither necessary nor convenient to use a large domain-specific dataset for training. On the contrary, by using only a handful of example images from a single generative model, a CLIP-based detector exhibits surprising generalization ability and high robustness across different architectures, including recent commercial tools such as Dalle-3, Midjourney v5, and Firefly. We match the state-of-the-art (SoTA) on in-distribution data and significantly improve upon it in terms of generalization to out-of-distribution data (+6% AUC) and robustness to impaired/laundered data (+13%). Our project is available at https://grip-unina.github.io/ClipBased-SyntheticImageDetection/

The ability to detect manipulated visual content is becoming increasingly important in many application fields, given the rapid advances in image synthesis methods. Of particular concern is the possibility of modifying the content of medical images, altering the resulting diagnoses. Despite its relevance, this issue has received limited attention from the research community. One reason is the lack of large and curated datasets to use for development and benchmarking purposes. Here, we investigate this issue and propose M3Dsynth, a large dataset of manipulated Computed Tomography (CT) lung images. We create manipulated images by injecting or removing lung cancer nodules in real CT scans, using three different methods based on Generative Adversarial Networks (GAN) or Diffusion Models (DM), for a total of 8,577 manipulated samples. Experiments show that these images easily fool automated diagnostic tools. We also tested several state-of-the-art forensic detectors and demonstrated that, once trained on the proposed dataset, they are able to accurately detect and localize manipulated synthetic content, even when training and test sets are not aligned, showing good generalization ability. Dataset and code are publicly available at https://grip-unina.github.io/M3Dsynth/.

This work presents a new unsupervised framework for training deep learning models for super-resolution of Sentinel-2 images by fusion of its 10-m and 20-m bands. The proposed scheme avoids the resolution downgrade process needed to generate training data in the supervised case. On the other hand, a proper loss that accounts for cycle-consistency between the network prediction and the input components to be fused is proposed. Despite its unsupervised nature, in our preliminary experiments the proposed scheme has shown promising results in comparison to the supervised approach. Besides, by construction of the proposed loss, the resulting trained network can be ascribed to the class of multi-resolution analysis methods.

In latest years, deep learning has gained a leading role in the pansharpening of multiresolution images. Given the lack of ground truth data, most deep learning-based methods carry out supervised training in a reduced-resolution domain. However, models trained on downsized images tend to perform poorly on high-resolution target images. For this reason, several research groups are now turning to unsupervised training in the full-resolution domain, through the definition of appropriate loss functions and training paradigms. In this context, we have recently proposed a full-resolution training framework which can be applied to many existing architectures. Here, we propose a new deep learning-based pansharpening model that fully exploits the potential of this approach and provides cutting-edge performance. Besides architectural improvements with respect to previous work, such as the use of residual attention modules, the proposed model features a novel loss function that jointly promotes the spectral and spatial quality of the pansharpened data. In addition, thanks to a new fine-tuning strategy, it improves inference-time adaptation to target images. Experiments on a large variety of test images, performed in challenging scenarios, demonstrate that the proposed method compares favorably with the state of the art both in terms of numerical results and visual output. Code is available online at https://github.com/matciotola/Lambda-PNN.

Detecting fake images is becoming a major goal of computer vision. This need is becoming more and more pressing with the continuous improvement of synthesis methods based on Generative Adversarial Networks (GAN), and even more with the appearance of powerful methods based on Diffusion Models (DM). Towards this end, it is important to gain insight into which image features better discriminate fake images from real ones. In this paper we report on our systematic study of a large number of image generators of different families, aimed at discovering the most forensically relevant characteristics of real and generated images. Our experiments provide a number of interesting observations and shed light on some intriguing properties of synthetic images: (1) not only the GAN models but also the DM and VQ-GAN (Vector Quantized Generative Adversarial Networks) models give rise to visible artifacts in the Fourier domain and exhibit anomalous regular patterns in the autocorrelation; (2) when the dataset used to train the model lacks sufficient variety, its biases can be transferred to the generated images; (3) synthetic and real images exhibit significant differences in the mid-high frequency signal content, observable in their radial and angular spectral power distributions.

Over the past decade, there has been tremendous progress in creating synthetic media, mainly thanks to the development of powerful methods based on generative adversarial networks (GAN). Very recently, methods based on diffusion models (DM) have been gaining the spotlight. In addition to providing an impressive level of photorealism, they enable the creation of text-based visual content, opening up new and exciting opportunities in many different application fields, from arts to video games. On the other hand, this property is an additional asset in the hands of malicious users, who can generate and distribute fake media perfectly adapted to their attacks, posing new challenges to the media forensic community. With this work, we seek to understand how difficult it is to distinguish synthetic images generated by diffusion models from pristine ones and whether current state-of-the-art detectors are suitable for the task. To this end, first we expose the forensics traces left by diffusion models, then study how current detectors, developed for GAN-generated images, perform on these new synthetic images, especially in challenging social-networks scenarios involving image compression and resizing. Datasets and code are available at github.com/grip-unina/DMimageDetection.

Thanks to recent advances in deep learning, sophisticated generation tools exist, nowadays, that produce extremely realistic synthetic speech. However, malicious uses of such tools are possible and likely, posing a serious threat to our society. Hence, synthetic voice detection has become a pressing research topic, and a large variety of detection methods have been recently proposed. Unfortunately, they hardly generalize to synthetic audios generated by tools never seen in the training phase, which makes them unfit to face real-world scenarios. In this work, we aim at overcoming this issue by proposing a new detection approach that leverages only the biometric characteristics of the speaker, with no reference to specific manipulations. Since the detector is trained only on real data, generalization is automatically ensured. The proposed approach can be implemented based on off-the-shelf speaker verification tools. We test several such solutions on three popular test sets, obtaining good performance, high generalization ability, and high robustness to audio impairment.

The ever higher quality and wide diffusion of fake images have spawn a quest for reliable forensic tools. Many GAN image detectors have been proposed, recently. In real world scenarios, however, most of them show limited robustness and generalization ability. Moreover, they often rely on side information not available at test time, that is, they are not universal. We investigate these problems and propose a new GAN image detector based on a limited sub-sampling architecture and a suitable contrastive learning paradigm. Experiments carried out in challenging conditions prove the proposed method to be a first step towards universal GAN image detection, ensuring also good robustness to common image impairments, and good generalization to unseen architectures.

In recent years, there has been a growing interest in deep learning-based pansharpening. Thus far, research has mainly focused on architectures. Nonetheless, model training is an equally important issue. A first problem is the absence of ground truths, unavoidable in pansharpening. This is often addressed by training networks in a reduced resolution domain and using the original data as ground truth, relying on an implicit scale invariance assumption. However, on full resolution images results are often disappointing, suggesting such invariance not to hold. A further problem is the scarcity of training data, which causes a limited generalization ability and a poor performance on off-training test images. In this paper, we propose a full-resolution training framework for deep learning-based pansharpening. The framework is fully general and can be used for any deep learning-based pansharpening model. Training takes place in the high-resolution domain, relying only on the original data, thus avoiding any loss of information. To ensure spectral and spatial fidelity, a suitable two-component loss is defined. The spectral component enforces consistency between the pansharpened output and the low-resolution multispectral input. The spatial component, computed at high-resolution, maximizes the local correlation between each pansharpened band and the panchromatic input. At testing time, the target-adaptive operating modality is adopted, achieving good generalization with a limited computational overhead. Experiments carried out on WorldView-3, WorldView-2, and GeoEye-1 images show that methods trained with the proposed framework guarantee a pretty good performance in terms of both full-resolution numerical indexes and visual quality.

The advent of deep learning has brought a significant improvement in the quality of generated media. However, with the increased level of photorealism, synthetic media are becoming hardly distinguishable from real ones, raising serious concerns about the spread of fake or manipulated information over the Internet. In this context, it is important to develop automated tools to reliably and timely detect synthetic media. In this work, we analyze the state-of-the-art methods for the detection of synthetic images, highlighting the key ingredients of the most successful approaches, and comparing their performance over existing generative architectures. We will devote special attention to realistic and challenging scenarios, like media uploaded on social networks or generated by new and unseen architectures, analyzing the impact of suitable augmentation and training strategies on the detectors' generalization ability., Comment: 7 pages, 5 figures, conference

Synthetic aperture radar (SAR) images are affected by a spatially-correlated and signal-dependent noise called speckle, which is very severe and may hinder image exploitation. Despeckling is an important task that aims at removing such noise, so as to improve the accuracy of all downstream image processing tasks. The first despeckling methods date back to the 1970's, and several model-based algorithms have been developed in the subsequent years. The field has received growing attention, sparkled by the availability of powerful deep learning models that have yielded excellent performance for inverse problems in image processing. This paper surveys the literature on deep learning methods applied to SAR despeckling, covering both the supervised and the more recent self-supervised approaches. We provide a critical analysis of existing methods with the objective to recognize the most promising research lines, to identify the factors that have limited the success of deep models, and to propose ways forward in an attempt to fully exploit the potential of deep learning for SAR despeckling.

Lung ultrasound imaging is reaching growing interest from the scientific community. On one side, thanks to its harmlessness and high descriptive power, this kind of diagnostic imaging has been largely adopted in sensitive applications, like the diagnosis and follow-up of preterm newborns in neonatal intensive care units. On the other side, state-of-the-art image analysis and pattern recognition approaches have recently proven their ability to fully exploit the rich information contained in these data, making them attractive for the research community. In this work, we present a thorough analysis of recent deep learning networks and training strategies carried out on a vast and challenging multicenter dataset comprising 87 patients with different diseases and gestational ages. These approaches are employed to assess the lung respiratory status from ultrasound images and are evaluated against a reference marker. The conducted analysis sheds some light on this problem by showing the critical points that can mislead the training procedure and proposes some adaptations to the specific data and task. The achieved results sensibly outperform those obtained by a previous work, which is based on textural features, and narrow the gap with the visual score predicted by the human experts., Comment: 7 pages

PRNU-based image processing is a key asset in digital multimedia forensics. It allows for reliable device identification and effective detection and localization of image forgeries, in very general conditions. However, performance impairs significantly in challenging conditions involving low quality and quantity of data. These include working on compressed and cropped images, or estimating the camera PRNU pattern based on only a few images. To boost the performance of PRNU-based analyses in such conditions we propose to leverage the image noiseprint, a recently proposed camera-model fingerprint that has proved effective for several forensic tasks. Numerical experiments on datasets widely used for source identification prove that the proposed method ensures a significant performance improvement in a wide range of challenging situations.

Due to limited computational and memory resources, current deep learning models accept only rather small images in input, calling for preliminary image resizing. This is not a problem for high-level vision problems, where discriminative features are barely affected by resizing. On the contrary, in image forensics, resizing tends to destroy precious high-frequency details, impacting heavily on performance. One can avoid resizing by means of patch-wise processing, at the cost of renouncing whole-image analysis. In this work, we propose a CNN-based image forgery detection framework which makes decisions based on full-resolution information gathered from the whole image. Thanks to gradient checkpointing, the framework is trainable end-to-end with limited memory resources and weak (image-level) supervision, allowing for the joint optimization of all parameters. Experiments on widespread image forensics datasets prove the good performance of the proposed approach, which largely outperforms all baselines and all reference methods., Comment: 13 pages, 12 figures, journal

Deep neural networks provide unprecedented performance in all image classification problems, taking advantage of huge amounts of data available for training. Recent studies, however, have shown their vulnerability to adversarial attacks, spawning an intense research effort in this field. With the aim of building better systems, new countermeasures and stronger attacks are proposed by the day. On the attacker's side, there is growing interest for the realistic black-box scenario, in which the user has no access to the neural network parameters. The problem is to design efficient attacks which mislead the neural network without compromising image quality. In this work, we propose to perform the black-box attack along a low-distortion path, so as to improve both the attack efficiency and the perceptual quality of the adversarial image. Numerical experiments on real-world systems prove the effectiveness of the proposed approach, both in benchmark classification tasks and in key applications in biometrics and forensics., Comment: 8 pages, journal

In the last few years, generative adversarial networks (GAN) have shown tremendous potential for a number of applications in computer vision and related fields. With the current pace of progress, it is a sure bet they will soon be able to generate high-quality images and videos, virtually indistinguishable from real ones. Unfortunately, realistic GAN-generated images pose serious threats to security, to begin with a possible flood of fake multimedia, and multimedia forensic countermeasures are in urgent need. In this work, we show that each GAN leaves its specific fingerprint in the images it generates, just like real-world cameras mark acquired images with traces of their photo-response non-uniformity pattern. Source identification experiments with several popular GANs show such fingerprints to represent a precious asset for forensic analyses.

We propose a new method for SAR image despeckling which leverages information drawn from co-registered optical imagery. Filtering is performed by plain patch-wise nonlocal means, operating exclusively on SAR data. However, the filtering weights are computed by taking into account also the optical guide, which is much cleaner than the SAR data, and hence more discriminative. To avoid injecting optical-domain information into the filtered image, a SAR-domain statistical test is preliminarily performed to reject right away any risky predictor. Experiments on two SAR-optical datasets prove the proposed method to suppress very effectively the speckle, preserving structural details, and without introducing visible filtering artifacts. Overall, the proposed method compares favourably with all state-of-the-art despeckling filters, and also with our own previous optical-guided filter.

With the ubiquitous diffusion of social networks, images are becoming a dominant and powerful communication channel. Not surprisingly, they are also increasingly subject to manipulations aimed at distorting information and spreading fake news. In recent years, the scientific community has devoted major efforts to contrast this menace, and many image forgery detectors have been proposed. Currently, due to the success of deep learning in many multimedia processing tasks, there is high interest towards CNN-based detectors, and early results are already very promising. Recent studies in computer vision, however, have shown CNNs to be highly vulnerable to adversarial attacks, small perturbations of the input data which drive the network towards erroneous classification. In this paper we analyze the vulnerability of CNN-based image forensics methods to adversarial attacks, considering several detectors and several types of attack, and testing performance on a wide range of common manipulations, both easily and hardly detectable.

Video forgery detection is becoming an important issue in recent years, because modern editing software provide powerful and easy-to-use tools to manipulate videos. In this paper we propose to perform detection by means of deep learning, with an architecture based on autoencoders and recurrent neural networks. A training phase on a few pristine frames allows the autoencoder to learn an intrinsic model of the source. Then, forged material is singled out as anomalous, as it does not fit the learned model, and is encoded with a large reconstruction error. Recursive networks, implemented with the long short-term memory model, are used to exploit temporal dependencies. Preliminary results on forged videos show the potential of this approach., Comment: Presented at IS&T Electronic Imaging: Media Watermarking, Security, and Forensics, January 2017

We propose a new algorithm for the reliable detection and localization of video copy-move forgeries. Discovering well crafted video copy-moves may be very difficult, especially when some uniform background is copied to occlude foreground objects. To reliably detect both additive and occlusive copy-moves we use a dense-field approach, with invariant features that guarantee robustness to several post-processing operations. To limit complexity, a suitable video-oriented version of PatchMatch is used, with a multiresolution search strategy, and a focus on volumes of interest. Performance assessment relies on a new dataset, designed ad hoc, with realistic copy-moves and a wide variety of challenging situations. Experimental results show the proposed method to detect and localize video copy-moves with good accuracy even in adverse conditions.

Local descriptors based on the image noise residual have proven extremely effective for a number of forensic applications, like forgery detection and localization. Nonetheless, motivated by promising results in computer vision, the focus of the research community is now shifting on deep learning. In this paper we show that a class of residual-based descriptors can be actually regarded as a simple constrained convolutional neural network (CNN). Then, by relaxing the constraints, and fine-tuning the net on a relatively small training set, we obtain a significant performance improvement with respect to the conventional detector.

We propose a new algorithm for fast approximate nearest neighbor search based on the properties of ordered vectors. Data vectors are classified based on the index and sign of their largest components, thereby partitioning the space in a number of cones centered in the origin. The query is itself classified, and the search starts from the selected cone and proceeds to neighboring ones. Overall, the proposed algorithm corresponds to locality sensitive hashing in the space of directions, with hashing based on the order of components. Thanks to the statistical features emerging through ordering, it deals very well with the challenging case of unstructured data, and is a valuable building block for more complex techniques dealing with structured data. Experiments on both simulated and real-world data prove the proposed algorithm to provide a state-of-the-art performance.

We explore the use of convolutional neural networks for the semantic classification of remote sensing scenes. Two recently proposed architectures, CaffeNet and GoogLeNet, are adopted, with three different learning modalities. Besides conventional training from scratch, we resort to pre-trained networks that are only fine-tuned on the target data, so as to avoid overfitting problems and reduce design time. Experiments on two remote sensing datasets, with markedly different characteristics, testify on the effectiveness and wide applicability of the proposed solution, which guarantees a significant performance improvement over all state-of-the-art references.