Swapping Algorithms Research Articles

Towards mitigating uncann(eye)ness in face swaps via gaze-centric loss terms

Advances in face swapping have enabled the automatic generation of highly realistic faces. Yet face swaps are perceived differently than when looking at real faces, with key differences in viewer behavior surrounding the eyes. Face swapping algorithms generally place no emphasis on the eyes, relying on pixel or feature matching losses that consider the entire face to guide the training process. We further investigate viewer perception of face swaps, focusing our analysis on the presence of an uncanny valley effect. We additionally propose a novel loss equation for the training of face swapping models, leveraging a pretrained gaze estimation network to directly improve representation of the eyes. We confirm that viewed face swaps do elicit uncanny responses from viewers. Our proposed improvements significant reduce viewing angle errors between face swaps and their source material. Our method additionally reduces the prevalence of the eyes as a deciding factor when viewers perform deepfake detection tasks. Our findings have implications on face swapping for special effects, as digital avatars, as privacy mechanisms, and more; negative responses from users could limit effectiveness in said applications. Our gaze improvements are a first step towards alleviating negative viewer perceptions via a targeted approach.

A Security-Aware and LUT-Based CAD Flow for the Physical Synthesis of hASICs

Numerous threats are associated with the globalized integrated circuit (IC) supply chain, such as piracy, reverse engineering, overproduction, and malicious logic insertion. Many obfuscation approaches have been proposed to mitigate these threats by preventing an adversary from fully understanding the IC (or parts of it). The use of reconfigurable elements inside an IC is a known obfuscation technique, either as a coarse grain reconfigurable block (i.e., eFPGA) or as a fine grain element (i.e., FPGA-like look-up tables). This paper presents a security-aware CAD flow that is LUT-based yet still compatible with the standard cell based physical synthesis flow. More precisely, our CAD flow explores the FPGA-ASIC design space and produces heavily obfuscated designs where only small portions of the logic resemble an ASIC. Therefore, we term this specialized solution a “hybrid ASIC” (hASIC). Nevertheless, even for heavily LUT-dominated designs, our proposed decomposition and pin swapping algorithms allow for performance gains that enable performance levels that only ASICs would otherwise achieve. On the security side, we have developed novel template-based attacks and also applied existing attacks, both oracle-free and oracle-based. Our security analysis revealed that the obfuscation rate for an SHA-256 study case should be at least 45% for withstanding traditional attacks and at least 80% for withstanding template-based attacks. When the 80% obfuscated SHA-256 design is physically implemented, it achieves a remarkable frequency of 368MHz in a 65nm commercial technology, whereas its FPGA implementation (in a superior technology) achieves only 77MHz.

Proposing a Simulation-Based Dynamic System Optimal Traffic Assignment Algorithm for SUMO: An Approximation of Marginal Travel Time

User equilibrium (UE) and system optimal (SO) are among the essential principles for solving the traffic assignment problem. Many studies have been performed on solving the UE and SO traffic assignment problem; however, the majority of them are either static (which can lead to inaccurate predictions due to long aggregation intervals) or analytical (which is computationally expensive for large-scale networks). Besides, most of the well-known micro/meso traffic simulators, do not provide a SO solution of the traffic assignment problem. To this end, this study proposes a new simulation-based dynamic system optimal (SB-DSO) traffic assignment algorithm for the SUMO simulator, which can be applied on large-scale networks. A new swapping/convergence algorithm, which is based on the logit route choice model, is presented in this study. This swapping algorithm is compared with the Method of Successive Average (MSA) which is very common in the literature. Also, a surrogate model of marginal travel time was implemented in the proposed algorithm, which was tested on real and abstract road networks (both on micro and meso scales). The results indicate that the proposed swapping algorithm has better performance than the classical swapping algorithms (e.g. MSA). Furthermore, a comparison was made between the proposed SB-DSO and the current simulation-based dynamic user equilibrium (SB-DUE) traffic assignment algorithm in SUMO. This proposed algorithm helps researchers to better understand the impacts of vehicles that may follow SO routines in future (e.g., Connected and Autonomous Vehicles (CAVs)).

Analyzing students’ experience in programming with computational thinking through competitive, physical, and tactile games: the quadrilateral method approach

The lack of computational thinking (CT) skills can be one of the reasons why students find themselves having difficulties in writing a good program. Therefore, understanding how CT skills can be developed is essential. This research explores how CT skills can be developed for programming through competitive, physical, and tactile games. The CT elements in this research focus on four major programming concepts, which are decomposition, pattern recognition, abstraction, and algorithmic thinking. We have conducted game activities through several algorithms that include sorting, swapping, and graph algorithms and analyzed how the game affects the student experience (SX) in understanding the CT concept in those algorithms. We have applied the quadrilateral method approach to the data collection and analysis. The data was obtained through observation, interview/survey based on six SX criteria (attention, engagement, awareness, satisfaction, confidence, and performance), and performances of the conducted game activities were compared. The results of the quadrilation of the data collected show a positive impact on the SX, highlight the effectiveness of the competitive, physical, and tactile game approach proposed in this research towards programming and CT skills development.

Joint User Association and Passive Beamforming in Heterogeneous Networks With Reconfigurable Intelligent Surfaces

In this letter, the joint design of user association (UA) and passive beamforming (PBF) has been investigated for heterogeneous networks equipped with reconfigurable intelligent surfaces. Firstly, a closed-form lower bound of ergodic rate is derived using statistical channel state information and adopting the maximum ratio transmission precoding. After that, a joint UA and PBF problem is formulated to maximize the weighted sum-rate (WSR), which is then decomposed into UA and PBF subproblems and solved by our proposed priority-based swapping and fixed-point iteration algorithms. Simulation results demonstrate that our proposed scheme achieves significant WSR performance gain when compared to the state-of-the-art methods.

Cross‐comparison of convergence algorithms to solve trip‐based dynamic traffic assignment problems

AbstractSolving a dynamic traffic assignment problem in a transportation network is a computational challenge. This study first reviews the different algorithms in the literature used to numerically calculate the user equilibrium (UE) related to dynamic network loading. Most of them are based on iterative methods to solve a fixed‐point problem. Two elements must be computed: the path set and the optimal path flow distribution between all origin–destination pairs. In a generic framework, these two steps are referred to as the outer and the inner loops, respectively. The goal of this study is to assess the computational performance of the inner loop methods that calculate the path flow distribution for different network settings (mainly network size and demand levels). Several improvements are also proposed to speed up convergence: four new swapping algorithms and two new methods for the step size initialization used in each descent iteration. All these extensions significantly reduce the number of iterations to obtain a good convergence rate and drastically speed up the overall simulations. The results show that the performance of different components of the solution algorithm is sensitive to the network size and saturation. Finally, the best algorithms and settings are identified for all network sizes with particular attention being given to the largest scale.

Face Swapping: Realistic Image Synthesis Based on Facial Landmarks Alignment

We propose an image-based face swapping algorithm, which can be used to replace the face in the reference image with the same facial shape and features as the input face. First, a face alignment is made based on a group of detected facial landmarks, so that the aligned input face and the reference face are consistent in size and posture. Secondly, an image warping algorithm based on triangulation is presented to adjust the reference face and its background according to the aligned input faces. In order to achieve more accurate face swapping, a face parsing algorithm is introduced to realize the accurate detection of the face-ROIs, and then the face-ROI in the reference image is replaced with the input face-ROI. Finally, a Poisson image editing algorithm is adopted to realize the boundary processing and color correction between the replacement region and the original background, and then the final face swapping result is obtained. In the experiments, we compare our method with other face swapping algorithms and make a qualitative and quantitative analysis to evaluate the reality and the fidelity of the replaced face. The analysis results show that our method has some advantages in the overall performance of swapping effect.

A self-regulating pairwise swapping algorithm to search reliability-based user equilibrium

The violation of monotonicity on reliability measures (RMs) usually makes the mathematical programming algorithms less efficient in solving the reliability-based user equilibrium (RUE) problem. The swapping algorithms provide a simple and convenient alternative to search traffic equilibrium since they are derivative-free and require weaker monotonicity. However, the existing swapping algorithms are usually based on linear swapping processes which cannot naturally avoid overswapping, and the step-size parameter update methods do not take the swapping feature into account. In this paper, we suggest a self-regulating pairwise swapping algorithm (SRPSA) to search RUE. SRPSA comprises an RM-based pairwise swapping process (RMPSP), a parameter self-diminishing operator and a termination criterion. SRPSA does not need to check the feasibility of either solutions or step-size parameter. It is suggested from the numerical analyses that SRPSA is effective and can swap to the quasi-RUE very fast. Therefore, SRPSA offers a good approach to generate initial points for those superior local search algorithms.

Global-Searching-Based Iterative Swapping Antenna Selection for Massive MIMO Systems with Imperfect Channel Estimation

With the growing of base station antenna number in massive multiple-input-multiple-output (MIMO), the complexity and the cost of the system rise rapidly. Fortunately, the antenna selection is an effective way to solve the problem, which can capture most of the advantages of massive MIMO. In this paper, a novel global-searching-based iterative swapping antenna selection method is proposed under the partial channel knowledge case for massive MIMO based on capacity maximization. The method is investigated with detailed mathematical derivations, as well as convergence discussion. Both global-searching-based “local” swapping (GSL-swapping) and global-searching-based “global” swapping algorithms are proposed for massive MIMO with imperfect channel state information (CSI). The former exchanges the selected antenna with the unselected one which is “better” than the selected one. Different from the GSL-swapping algorithm, the latter only swaps the “worst” selected antenna with the “best” unselected antenna which is “better” than the “worst” selected one. In addition, an improved global-searching-based “global” swapping algorithm with imperfect CSI is proposed to further decrease the complexity. Due to the “global searching” characteristic, the proposed algorithms can obtain a near optimal performance. Numerical simulations are provided to validate the proposed algorithms.

A Large Deviations Analysis of Certain Qualitative Properties of Parallel Tempering and Infinite Swapping Algorithms

Parallel tempering, or replica exchange, is a popular method for simulating complex systems. The idea is to run parallel simulations at different temperatures, and at a given swap rate exchange configurations between the parallel simulations. From the perspective of large deviations it is optimal to let the swap rate tend to infinity and it is possible to construct a corresponding simulation scheme, known as infinite swapping. In this paper we propose a novel use of large deviations for empirical measures for a more detailed analysis of the infinite swapping limit in the setting of continuous time jump Markov processes. Using the large deviations rate function and associated stochastic control problems we consider a diagnostic based on temperature assignments, which can be easily computed during a simulation. We show that the convergence of this diagnostic to its a priori known limit is a necessary condition for the convergence of infinite swapping. The rate function is also used to investigate the impact of asymmetries in the underlying potential landscape, and where in the state space poor sampling is most likely to occur.

Formula omitted]-Optimal supersaturated designs with good minimax properties

An improved E ( s 2 ) lower bound is derived for two-level supersaturated designs. This improved bound is used to prove E ( s 2 ) -optimality of the best designs obtained via algorithmic search in all cases with N = 10 , 12, 14, and 16 runs (except the N = 14 run, m = 16 factor case). New exchange algorithms which generalize the NOA algorithm of Nguyen [1996. An algorithmic approach to constructing supersaturated designs. Technometrics 38, 69–73] and which tend to find E ( s 2 ) -optimal designs with better minimax properties are proposed. Row swapping algorithms are used to find E ( s 2 ) -optimal designs when the number of factors is large. E ( s 2 ) -optimal designs found via algorithmic search are compared to cyclicly constructed E ( s 2 ) -optimal designs using the minimax criterion.

On swapping and simulated tempering algorithms

In this paper we study the relationships between two Markov Chain Monte Carlo algorithms—the Swapping Algorithm (also known as the Metropolis-coupled algorithm) and the simulated tempering algorithm. We give a proof that the spectral gap of the simulated tempering chain is bounded below by a multiple of that of the swapping chain.

Optimisation of synthesised array excitations using array polynome complex root swapping and genetic algorithms

A numerical pattern synthesis and optimisation technique for shaped and low sidelobe beams adhering to desired array element excitations are described. The synthesis approach is a modification of the Woodward synthesis method with an additional sidelobe minimisation, and the final element excitation optimisation is based on array polynome complex root swapping and genetic algorithms. These methods have been applied to an equispaced linear planar array. Results of the synthesis and genetic algorithms are presented.

Factors Affecting the Efficiency of A Virtual Memory

This paper compares empirical results that have been obtained for several virtual memory swapping algorithms. The algorithms were tested as software components of a multiple computer on-line system.