The aim of this article is to evaluate a state of the art image synthesis carried out via Generative Adversarial Networks (conditional Wasserstein GAN and Self Attention GAN) on a traffic signs dataset. For the experiment, we focused on generating images with a 64×64-pixel res...

We study how well different types of approaches generalise in the task of 3D hand pose estimation under single hand scenarios and hand-object interaction. We show that the accuracy of state-of-the-art methods can drop, and that they fail mostly on poses absent from the trainin...

In this paper, we propose a pipeline for processing of scanned historical documents into the electronic text form that could then be indexed and stored in a database. The nature of the documents presents a substantial challenge for standard automated techniques — not only ther...

In this paper, we present our experiments with BERT models in the task of Large-scale Multi-label Text Classification (LMTC). In the LMTC task, each text document can have multiple class labels, while the total number of classes is in the order of thousands. We propose a&...

Phrase boundary detection is an important part of text-to-speech systems since it ensures more natural speech synthesis outputs. However, the problem of phrasing is ambiguous, especially per speaker and per style. This is the reason why this paper focuses on speaker-dependent phrasing f...

While unit selection speech synthesis is not at the centre of research nowadays, it shows its strengths in deployments where fast fixes and tuning possibilities are required. The key part of this method is target and concatenation costs, usually consisting of features manually desi...

Modern web browsers are becoming operating systems of their own kind, allowing unified access to the underlying hardware. The sound device can thus be used by web-based communication systems, such a Google meet, Zoom and others. This attracts the idea of using such capabilities...

In this paper, we discuss some of the properties of training acoustic models using a lattice-free version of the maximum mutual information criterion (LF-MMI). Currently, the LF-MMI method achieves state-of-the-art results on many speech recognition tasks. Some of the key features of th...

This paper is devoted to the problem of solving the steady incompressible Navier–Stokes equations discretized by the Galerkin method on the spaces generated by the B-spline/NURBS basis functions, which is called isogeometric analysis. Two pressure-correction methods are presented for the sol...

We construct a polynomial-time algorithm that, for a graph X with 4p We construct a polynomial-time algorithm that for a graph X with 4p vertices (p is prime), finds (if any) a Cayley representation of X over the group C2 × C2 × Cp. This result, together with&#x...

Linear fusion of estimates is a basic tool for combining probabilistic data. If the correlation of estimation errors is unknown, the fusion performance is evaluated with respect to the worst case. Inverse Covariance Intersection fusion is a rule for combining two estimates with par...

This paper deals with the state estimation of nonlinear stochastic dynamic systems, where the measurement noise is modelled by the multimodal Gaussian sum probability density function. The multimodal density is able to sufficiently capture various environmental features and phenomena affecting se...

The paper deals with the active fault diagnosis of large scale stochastic systems with faults modeled as mutually dependent Markov chains. The system is described by multiple models representing fault-free and faulty behavior of the system. The aim of the active fault detector in&#...

The paper deals with the state estimation of nonlinear stochastic systems with additive Gaussian noises by means of the Gaussian filters leveraging numerical integration rules. The filters were derived under the assumption of the joint state and measurement predictive density being Gaussian,...

Constrained state estimation has been studied in the literature usually from the point estimation perspective. That is, a random vector is designed and its covariance matrix is searched based on contextual information available in addition to the measured data. This paper deals with...

The paper is devoted to the state estimation of nonlinear stochastic dynamic systems. The stress is laid on a grid-based numerical solution to the Bayesian recursive relations using the point-mass filter (PMF). In the paper, a novel conditional density driven grid (CDDG) design is&...

The nonlinear second order ordinary differential equation u''(x) + m^2 u(x) + g(x, u(x)) = f(x) , x in [0, p] , u(0) = u(p) = 0 is investigated in this article. We suppose that the nonlinearity g is unbounded and satisfies an ALP type condition. We prove...

One of the applications of ULS (UAV-borne laser scanning) lies in power line inspection. However, with LiDAR data (i.e. point clouds) comes the need for reliable automatic classification, also called semantic segmentation, of data which allows further analysis of gathered data. Vast num...

Thispaperdealswiththeexperimentalandnumericalinvestigationsofmechanicalbehavior of the single-lap joint of laminates consisting of TENAX-J IMS60 E13 24K carbon fibers and MSG L285 epoxy resin. TIBULB high-strength blind rivets having a diameter of 4.8 mm and 6.4 mm were used for the joint. Tension tes...

The purpose of our work is to develop an automatic shape optimization tool for runner wheel blades in reaction water turbines, especially in Kaplan turbines. The fluid flow is simulated using an in-house incompressible turbulent flow solver based on recently introduced isogeometric anal...