The revolution in computer hardware, especially in graphics processing units and tensor processing units, has enabled significant advances in computer graphics and artificial intelligence algorithms. In addition to their many beneficial applications in daily life and business, computergenerated/manipulated images and videos can be used for malicious purposes that violate security systems, privacy, and social trust. The deepfake phenomenon and its variations enable a normal user to use his or her personal computer to easily create fake videos of anybody from a short real online video. Several countermeasures have been introduced to deal with attacks using such videos. However, most of them are targeted at certain domains and are ineffective when applied to other domains or new attacks. In this paper, we introduce a capsule network that can detect various kinds of attacks, from presentation attacks using printed images and replayed videos to attacks using fake videos created using deep learning. It uses many fewer parameters than traditional convolutional neural networks with similar performance. Moreover, we explain, for the first time ever in the literature, the theory behind the application of capsule networks to the forensics problem through detailed analysis and visualization.

Detecting manipulated images and videos is an important topic in digital media forensics. Most detection methods use binary classification to determine the probability of a query being manipulated. Another important topic is locating manipulated regions (i.e., performing segmentation), which are mostly created by three commonly used attacks: removal, copy-move, and splicing. We have designed a convolutional neural network that uses the multi-task learning approach to simultaneously detect manipulated images and videos and locate the manipulated regions for each query. Information gained by performing one task is shared with the other task and thereby enhance the performance of both tasks. A semi-supervised learning approach is used to improve the network's generability. The network includes an encoder and a Y-shaped decoder. Activation of the encoded features is used for the binary classification. The output of one branch of the decoder is used for segmenting the manipulated regions while that of the other branch is used for reconstructing the input, which helps improve overall performance. Experiments using the FaceForensics and Face-Forensics++ databases demonstrated the networks effectiveness against facial reenactment attacks and face swapping attacks as well as its ability to deal with the mismatch condition for previously seen attacks. Moreover, fine-tuning using just a small amount of data enables the network to deal with unseen attacks.

No abstract

Electrospinning has received a lot of attention in recent years as a simple process to produce sub-micrometer-scale fibers. The process involves continuous stretching of polymer solution or melt in the presence of a strong electric field, which forms ultrathin fibers. A large amount of research is being carried out to achieve control of the diameter, morphology, and spatial alignment of electrospun nanofibers. [1,2] Unlike other 1D nanostructures, such as nanowires and nanotubes, nanofibers exhibit a wide range of unique properties, making them far more attractive for many applications, such as filtration, catalysis, sensing, protective clothing, and tissueengineering scaffolds.[3] Being a continuous process, electrospinning can produce extremely long fibers comparable to those formed via conventional mechanical drawing and spinning techniques. Nanofiber mats exhibit extremely high surface-to-mass ratios, greatly improving their efficiency for catalysis and filtration. [2] Another fascinating feature of this process is that it can be applied to a wide variety of nanostructured materials. Bognitzki et al. [4] fabricated sub-micrometerscale fibers by electrospinning ternary solutions of polylactide and poly(vinylpyrrolidone). They found that the fibers displayed an internal phase morphology, and, by selectively removing one phase, they obtained highly porous fibers. However, these morphologies were largely controlled by rapid phase separation and rapid solidification of the polymer jet, and no method for controlling these morphologies was demonstrated. In the present study, we utilize this simple process to fabricate block copolymer (BCP) nanofibers. BCP solutions and melts are known to self-assemble into a variety of nanoscale morphologies including spheres, rods, micelles, lamellae, vesicles, tubules, and cylinders, [5,6] depending on the volume fraction and interaction parameter between different blocks. We have been able to produce macroscale-length fibers with diameters of a few hundred nanometers that exhibit internal structures of only tens of nanometers in size. Such materials, we believe, can combine the unique properties of continuous nanofibers and BCP self-assembly for use in a variety of applications of nanostructured materials. BCP self-assembly has attracted increasing interest in recent years for applications in nanotechnology.[7] Precise control over the size, shape, and periodicity of these nanoscale microdomains is the key factor needed to realize nanoscale systems. Various methods, including shear and elongational deformation, compressional deformation, electric fields, and temperature gradients, have been utilized to induce orientation of the microdomains. To our knowledge, shear flow has been most extensively studied as a simple means to induce phase transitions and orient self-assembled structures in block copolymers. [8,9] The phenomenon of flow-induced alignment of lamellar BCPs is very well studied in bulk systems. [10][11][12] Three different orientations of lamellar morphology, namely, p...

No abstract