Search

The tremendous success of deep learning for imaging applications has resulted in numerous beneficial advances. Unfortunately, this success has also been a catalyst for malicious uses such as photo-realistic face swapping of parties without consent. Transferring one person's face from a source image to a target image of another person, while keeping the image photo-realistic overall has become increasingly easy and automatic, even for individuals without much knowledge of image processing. In this study, we use deep transfer learning for face swapping detection, showing true positive rates >96% with very few false alarms. Distinguished from existing methods that only provide detection accuracy, we also provide uncertainty for each prediction, which is critical for trust in the deployment of such detection systems. Moreover, we provide a comparison to human subjects. To capture human recognition performance, we build a website to collect pairwise comparisons of images from human subjects. Based on these comparisons, images are ranked from most real to most fake. We compare this ranking to the outputs from our automatic model, showing good, but imperfect, correspondence with linear correlations >0.75. Overall, the results show the effectiveness of our method. As part of this study, we create a novel, publicly available dataset that is, to the best of our knowledge, the largest public swapped face dataset created using still images. Our goal of this study is to inspire more research in the field of image forensics through the creation of a public dataset and initial analysis., Comment: 8 pages, 5 figures

Vorticity distributions in axisymmetric vortex rings produced by a piston-pipe apparatus are numerically studied over a range of Reynolds numbers, $\mathrm{Re}$, and stroke-to-diameter ratios, $L/D$. It is found that a state of advective balance, such that $\zeta \equiv \omega_\phi/r \approx F(\psi, t)$, is achieved within the region (called the vortex ring bubble) enclosed by the dividing streamline. Here $\zeta \equiv\omega_\phi/r$ is the ratio of azimuthal vorticity to cylindrical radius, and $\psi$ is the Stokes streamfunction in the frame of the ring. Some but not all of the $\mathrm{Re}$ dependence in the time evolution of $F(\psi, t)$ can be captured by introducing a scaled time $\tau = \nu t$, where $\nu$ is the kinematic viscosity. When $\nu t/D^2 \gtrsim 0.02$, the shape of $F(\psi)$ is dominated by the linear-in-$\psi$ component, the coefficient of the quadratic term being an order of magnitude smaller. An important feature is that as the dividing streamline ($\psi = 0$) is approached, $F(\psi)$ tends to a non-zero intercept which exhibits an extra $\mathrm{Re}$ dependence. This and other features are explained by a simple toy model consisting of the one-dimensional cylindrical diffusion equation. The key ingredient in the model responsible for the extra $\mathrm{Re}$ dependence is a Robin-type boundary condition, similar to Newton's law of cooling, that accounts for the edge layer at the dividing streamline.

The efficient use of limited computational resources is an essential ingredient of intelligence. Selecting computations optimally according to rational metareasoning would achieve this, but this is computationally intractable. Inspired by psychology and neuroscience, we propose the first concrete and domain-general learning algorithm for approximating the optimal selection of computations: Bayesian metalevel policy search (BMPS). We derive this general, sample-efficient search algorithm for a computation-selecting metalevel policy based on the insight that the value of information lies between the myopic value of information and the value of perfect information. We evaluate BMPS on three increasingly difficult metareasoning problems: when to terminate computation, how to allocate computation between competing options, and planning. Across all three domains, BMPS achieved near-optimal performance and compared favorably to previously proposed metareasoning heuristics. Finally, we demonstrate the practical utility of BMPS in an emergency management scenario, even accounting for the overhead of metareasoning.