From its acquisition in the camera sensors to its storage, different operations are performed to generate the final image. This pipeline imprints specific traces into the image to form a natural watermark. Tampering with an image disturbs these traces; these disruptions are clues that are used by most methods to detect and locate forgeries. In this article, we assess the capabilities of diffusion models to erase the traces left by forgers and, therefore, deceive forensics methods. Such an approach has been recently introduced for adversarial purification, achieving significant performance. We show that diffusion purification methods are well suited for counter-forensics tasks. Such approaches outperform already existing counter-forensics techniques both in deceiving forensics methods and in preserving the natural look of the purified images. The source code is publicly available at https://github.com/mtailanian/diff-cf.

Anomalies can be defined as any non-random structure which deviates from normality. Anomaly detection methods reported in the literature are numerous and diverse, as what is considered anomalous usually varies depending on particular scenarios and applications. In this work we propose an a contrario framework to detect anomalies in images applying statistical analysis to feature maps obtained via convolutions. We evaluate filters learned from the image under analysis via patch PCA and the feature maps obtained from a pre-trained deep neural network (Resnet). The proposed method is multi-scale and fully unsupervised, and is able to detect anomalies in a wide variety of scenarios. While the end goal of this work is the detection of subtle defects in leather samples for the automotive industry, we show that the same algorithm achieves state-of-the-art results in public anomalies datasets.

Web applications are permanently being exposed to attacks that exploit their vulnerabilities. In this work we investigate the application of machine learning techniques to leverage Web Application Firewalls (WAF)s, a technology that is used to detect and prevent attacks. We put forward an approach of complementary machine learning models, based on one-class classification and n-gram analysis, to enhance the detection and accuracy capabilities of MODSECURITY, an open source and widely used WAF. The results are promising and outperform MODSECURITY when configured with the OWASP Core Rule Set, the baseline configuration setting of a widely deployed, rule-based WAF technology.

Web applications are permanently being exposed to attacks that exploit their vulnerabilities. In this work we investigate the use of machine learning techniques to leverage the performance of Web Application Firewalls (WAFs), systems that are used to detect and prevent attacks. We propose a characterization of the problem by defining different scenarios depending if we have valid and/or attack data available for training. We also propose two solutions: first a multi-class approach for the scenario when valid and attack data is available; and second a one-class solution when only valid data is at hand. We present results using both approaches that outperform MODSECURITY configured with the OWASP Core Rule Set out of the box, which is the baseline configuration setting of a widely deployed WAF technology. We also propose a tagged dataset based on the DRUPAL content management framework.

Shape models (SMs), capturing the common features of a set of training shapes, represent a new incoming object based on its projection onto the corresponding model. Given a set of learned SMs representing different objects classes, and an image with a new shape, this work introduces a joint classification-segmentation framework with a twofold goal. First, to automatically select the SM that best represents the object, and second, to accurately segment the image taking into account both the image information and the features and variations learned from the online selected model. A new energy functional is introduced that simultaneously accomplishes both goals. Model selection is performed based on a shape similarity measure, online determining which model to use at each iteration of the steepest descent minimization, allowing for model switching and adaptation to the data. High-order SMs are used in order to deal with very similar object classes and natural variability within them. Position and transformation invariance is included as part of the modeling as well. The presentation of the framework is complemented with examples for the difficult task of simultaneously classifying and segmenting closely related shapes, such as stages of human activities, in images with severe occlusions.

Shape models (SMs), capturing the common features of a set of training shapes, represent a new incoming object based on its projection onto the corresponding model. Given a set of learned SMs representing different objects, and an image with a new shape, this work introduces a joint classification-segmentation framework with a twofold goal. First, to automatically select the SM that best represents the object, and second, to accurately segment the image taking into account both the image information and the features and variations learned from the on-line selected model. A new energy functional is introduced that simultaneously accomplishes both goals. Model selection is performed based on a shape similarity measure, determining which model to use at each iteration of the steepest descent minimization, allowing for model switching and adaptation to the data. High-order SMs are used in order to deal with very similar object classes and natural variability within them. The presentation of the framework is complemented with examples for the difficult task of simultaneously classifying and segmenting closely related shapes, stages of human activities, in images with severe occlusions.

Here, we present an efficient method for movie denoising that does not require any motion estimation. The method is based on the well-known fact that averaging several realizations of a random variable reduces the variance. For each pixel to be denoised, we look for close similar samples along the level surface passing through it. With these similar samples, we estimate the denoised pixel. The method to find close similar samples is done via warping lines in spatiotemporal neighborhoods. For that end, we present an algorithm based on a method for epipolar line matching in stereo pairs which has per-line complexity O (N), where N is the number of columns in the image. In this way, when applied to the image sequence, our algorithm is computationally efficient, having a complexity of the order of the total number of pixels. Furthermore, we show that the presented method is unsupervised and is adapted to denoise image sequences with an additive white noise while respecting the visual details on the movie frames. We have also experimented with other types of noise with satisfactory results.

Minimal surface regularization has been used in several applications ranging from stereo to image segmentation, sometimes hidden as a graph-cut discrete formulation, or as a strictly convex approximation to TV minimization. In this paper we consider a modified version of minimal surface regularization coupled with a robust data fitting term for interpolation purposes, where the corresponding evolution equation is constrained to diffuse only along the isophotes of a given image u and we design a convergent numerical scheme to accomplish this. To illustrate the usefulness of our approach, we apply this framework to the digital elevation model interpolation and to constrained vector probability diffusion. 1

A novel paradigm for information visualization in high dynamic range images is presented in this paper. These images, real or synthetic, have luminance with typical ranges many orders of magnitude higher than that of standard output/viewing devices, thereby requiring some processing for their visualization. In contrast with existent approaches, which compute a single image with reduced range, close in a given sense to the original data, we propose to look for a representative set of images. The goal is then to produce a minimal set of images capturing the information all over the high dynamic range data, while at the same time preserving a natural appearance for each one of the images in the set. A specific algorithm that achieves this goal is presented and tested on natural and synthetic data.

In this work, we study the extraction of semantic objects using morphological tools. We decompose the image into its level sets and level lines (the borders of the level sets). Specifically, from all the level lines we extract the ones that contain T-junctions, have compact form, and are well-contrasted to obtain the semantic objects in the scene. For all these factors, we establish an estimation procedure.

In this work, we study the extraction of semantic objects from still images. We combine different ideas to extract them in a structured manner together with a perceptual metric that ranks them according with its perceptual relevance. The algorithm has four steps, the regularization of the initial segmentation using probability diffusion, simplification of the segmentation via region merging, computation of the perceptual metric and construction of the structure that represents the image (the binary partition tree).

A novel paradigm for information visualization in high dynamic range images is presented. These images, real or synthetic, have luminance with typical ranges many orders of magnitude higher than that of standard output devices, thereby requiring some processing for their visualization. In contrast to existing approaches, which compute a single image with reduced range, close in a given sense to the original data, we propose to look for a representative set of images. The goal is then to produce a minimal set of images capturing the information all over the high dynamic range data, while at the same time preserving a natural appearance for each one of the images in the set. A specific algorithm that achieves this goal is presented and tested on natural and synthetic data.

The basic motivation of this work is to introduce contextual information into image segmentation tasks by adding spatial coherence to the posterior probabilities corresponding to the classes present in the scene. A method for isotropic and anisotropic diffusion of vector probabilities in general, and posterior probabilities in particular, is introduced. The technique is based on diffusing via coupled partial differential equations restricted to the semi-hyperplane corresponding to probability functions. Both the partial differential equations and their corresponding numerical implementation guarantee that the vector remains a probability vector, having all its components positive and adding to one. Applying the method to posterior probabilities in classification problems, spatial and contextual coherence is introduced before the maximum a posteriori (MAP) decision, thereby improving the classification results.

A method for isotropic and anisotropic diffusion of vector probabilities in general, and posterior probabilities in particular, is introduced. The technique is based on diffusing via coupled partial differential equations restricted to the semi-hyperplane corresponding to probability functions. Both the partial differential equations and their corresponding numerical implementation guarantee that the vector remains a probability vector, having all its components positive and adding to one. Applying the method to posterior probabilities in classification problems, spatial and contextual coherence is introduced before the MAP decision, thereby improving the classification results.