This paper presents a method to automatically and efficiently detect face tampering in videos, and particularly focuses on two recent techniques used to generate hyperrealistic forged videos: Deepfake and Face2Face. Traditional image forensics techniques are usually not well suited to videos due to the compression that strongly degrades the data. Thus, this paper follows a deep learning approach and presents two networks, both with a low number of layers to focus on the mesoscopic properties of images. We evaluate those fast networks on both an existing dataset and a dataset we have constituted from online videos. The tests demonstrate a very successful detection rate with more than 98% for Deepfake and 95% for Face2Face.

This paper presents a deep-learning method for distinguishing computer generated graphics from real photographic images. The proposed method uses a Convolutional Neural Network (CNN) with a custom pooling layer to optimize current best-performing algorithms feature extraction scheme. Local estimates of class probabilities are computed and aggregated to predict the label of the whole picture. We evaluate our work on recent photo-realistic computer graphics and show that it outperforms state of the art methods for both local and full image classification.

Discriminating between computer-generated images (CGIs) and photographic images (PIs) is not a new problem in digital image forensics. However, with advances in rendering techniques supported by strong hardware and in generative adversarial networks, CGIs are becoming indistinguishable from PIs in both human and computer perception. This means that malicious actors can use CGIs for spoofing facial authentication systems, impersonating other people, and creating fake news to be spread on social networks. The methods developed for discriminating between CGIs and PIs quickly become outdated and must be regularly enhanced to be able to reduce these attack surfaces. Leveraging recent advances in deep convolutional networks, we have built a modular CGI-PI discriminator with a customized VGG-19 network as the feature extractor, statistical convolutional neural networks as the feature transformers, and a discriminator. We also devised a probabilistic patch aggregation strategy to deal

Thermal imaging has become a valuable tool in various fields for remote sensing and can provide relevant information to perform object recognition or classification. In this paper, we present an automated method to obtain a 3D model fusing data from a visible and a thermal camera. The RGB and thermal point clouds are generated independently by structure from motion. The registration process includes a normalization of the point cloud scale, a global registration based on calibration data and the output of the structure from motion, and a fine registration employing a variant of the Iterative Closest Point optimization. Experimental results demonstrate the accuracy and robustness of the overall process.

Geometric Algebra can be understood as a set of tools to represent, construct and transform geometric objects. Some Geometric Algebras like the well-defined Conformal Geometric Algebra constructs lines, circles, planes, and spheres from control points just by using the outer product. There exist some Geometric Algebras to handle more complex objects such as quadric surfaces; however in this case, none of them is known to build quadric surfaces from control points. This paper presents a novel Geometric Algebra framework, the Geometric Algebra of R 9,6 , to deal with quadric surfaces where an arbitrary quadric surface is constructed by the mere wedge of nine points. The proposed framework enables us not only to intuitively represent quadric surfaces but also to construct objects using Conformal Geometric Algebra. Our proposed framework also provides the computation of the intersection of quadric surfaces, the normal vector, and the tangent plane of a quadric surface.

International audienceNoise is an intrinsic specificity of all forms of imaging, and can be found in various forms in all domains of digital imagery. This paper offers an overall review of digital image noise, from its causes and models to the degradations it suffers along the image acquisition pipeline. We show that by the end of the pipeline, the noise may have widely different characteristics compared to the raw image, and consider the consequences in forensic and counter-forensic imagery

We present a novel 3D display that can show any 3D contents in free space using laser-plasma scanning in the air. The laser-plasma technology can generate a point illumination at an arbitrary position in the free space. By scanning the position of the illumination, we can display a set of point illuminations in the space, which realizes 3D display in the space. This 3D display has been already presented in Emerging Technology of SIGGRAPH2006, which is the basic platform of our 3D display project. In this presentation, we would like to introduce history of the development of the laser-plasma scanning 3D display, and then describe recent development of the 3D contents analysis and processing technology for realizing an innovative media presentation in a free 3D space. The one of recent development is performed to give preferred 3D contents data to the 3D display in a very flexible manner. This means that we have a platform to develop an interactive 3D contents presentation system using the 3D display, such as an interactive art presentation using the 3D display. We would also like to present the future plan of this 3D display research project.

This paper presents both a recursive scheme to perform Geometric Algebra operations over a prefix tree, and Garamon, a C++ library generator implementing these recursive operations. While for low dimension vector spaces, precomputing all the Geometric Algebra products is an efficient strategy, it fails for higher dimensions where the operation should be computed at run time. This paper describes how a prefix tree can be a support for a recursive formulation of Geometric Algebra operations. This recursive approach presents a much better complexity than the usual run time methods. This paper also details how a prefix tree can represent efficiently the dual of a multivector. These results constitute the foundations for Garamon, a C++ library generator synthesizing efficient C++ / Python libraries implementing Geometric Algebra in both low and higher dimensions, with any arbitrary metric. Garamon takes advantage of the prefix tree formulation to implement Geometric Algebra operations on high dimensions hardly accessible with state-of-the-art software implementations. Garamon is designed to produce easy to install, easy to use, effective and numerically stable libraries. The design of the libraries is based on a data structure using precomputed functions for low dimensions and a smooth transition to the new recursive products for higher dimensions.