The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Tree structures are commonly used to model relationships between body parts for articulated Human Pose Estimation (HPE). Tree structures can be used to model relationships among feature maps of joints in a structured learning framework using Convolutional Neural Networks (CNNs). This paper proposes new data-driven tree models for HPE. The data-driven tree structures were obtained using the Chow-Liu...
Motion estimation is an essential procedure in video coding and object tracing, but it always has a high computational load. Some low bit-depth motion estimation methods, such as one/two-bit transform or gray coding based methods, have lower complexity. However, one-bit transform based methods are sensitive to noise, whereas two-bit transform and gray coding methods use many bit planes or operations...
In this study, we proposed an analysis method of ElectroMyoGraphic (EMG) signals in order to diagnose and to identify neuromuscular pathologies (i.e.; myopathy and neuropathy). Analysis is performed fully automatically without expert assistance and without prior segmentation of muscle contractions. The method is based on Huang-Hilbert transform (HHT) which is a data-driven algorithm that decomposes...
The main goal of the work is to present some aspects in using time-frequency transforms, applied to the vibration signals and for change detection purposes, from signal processing point of view. The processed object is the time-frequency image (TFI), as result of the time-frequency transform. Two distributions are considered: Wigner-Ville and Choi-Williams, each with specific properties and matched...
Ste gano graphic systems are used for the transmission of hidden data in the original signal. The article describes the algorithm of the hidden data transmission using the speech signal as a carrier. The echo method is used for data embedding. In order to improve the decoding efficiency of embedded data, the procedure of voicing correction and mechanism of informed coding were developed and implemented...
A novel projection twin support vector machine (PTSVM), termed as NPTSVM, is presented in this paper for binary classification. Although this method determines two projection vectors using the same way as PTSVM, it has more advantages than existing PTSVMs. First, NPTSVM does not have to calculate inverse matrices during the learning process, which makes the training speed of NPTSVM be much faster...
The paper deals with the problem of stability during the solving of pattern recognition tasks from the point of view of transformation groups. It shows the possibility to avoid the necessity of regularization by using the geometric equaffine Lorentz transformation, exploiting as example the alpha-procedure.
Graph matching is the task of computing the resemblance of graphs. While in exact matching, a strict one-to-one correspondence should exist between two graphs or among their subgraphs, on the other hand in error-tolerant matching a strict correspondence is not necessary, and some similarity measure should be exhibited between two graphs or their subgraph depending on some tolerance value or noise...
Recently a discrete Hankel transform (DHT) has been introduced using a symmetric involutory kernel matrix T. Although Namias contributed the fractional Hankel transform (FRHT) in 1980, no discrete counterpart has appeared till now. Here a definition is proposed for a discrete fractional Hankel transform (DFRHT) based on the eigen decomposition of the diagonalizable matrix T. Being a real symmetric...
Multiple transforms have received considerable attention recently, especially in the course of an exploration conducted by MPEG and ITU toward the standardization of the next generation video compression algorithm. This joint team has developed a software, called the Joint Exploration Model (JEM) which outperforms by over 25% the HEVC standard. The transform step in JEM consists in Adaptive Multiple...
Automatic blobs detection constitutes a basic but difficult problem. In this work a new fast blobs detection technique based on a scale-space representation of the original image, is proposed. The scale-space representation is constructed by using a new simplified form of the Fast Radial Symmetry Transform to precisely detect the essential blobs. From the experiments we have conducted, the proposed...
Time-frequency representation (TFR), revealing time and frequency features of signals simultaneously, has attracted increasingly attention in non-stationary signal processing. The existing TFRs can be mainly divided into two categories: non-parameterized and parameterized TFRs. The former has been widely applied on various non-stationary signals, while the latter is of particular interesting since...
Scene flow describes the motion of 3D objects in real world and potentially could be the basis of a good feature for 3D action recognition. However, its use for action recognition, especially in the context of convolutional neural networks (ConvNets), has not been previously studied. In this paper, we propose the extraction and use of scene flow for action recognition from RGB-D data. Previous works...
The aim of this article is to design a moment transformation for Student-t distributed random variables, which is able to account for the error in the numerically computed mean. We employ Student-t process quadrature, an instance of Bayesian quadrature, which allows us to treat the integral itself as a random variable whose variance provides information about the incurred integration error. Advantage...
In this paper we propose a technique to deal with unknown random jitter in a band-limited Gaussian channel. The jitter caused by the deviation or displacement of signal pulses affects the performance of the communication system. We show that a sampling set as small as twice the baud rate is enough for good detection performance. Detection is done by means of a suboptimal algorithm with polynomial...
Existing GPU graph analytics frameworks are typically built from specialized, bottom-up implementations of graph operators that are customized to graph computation. In this work we describe Mini-Gunrock, a lightweight graph analytics framework on the GPU. Unlike existing frameworks, Mini-Gunrock is built from graph operators implemented with generic transform-based data-parallel primitives. Using...
In this paper, we propose a novel scheme for domain adaptation in which feature transform and instance weights are jointly optimized. Due to the joint optimization, we can obtain feasible feature transform for domain adaptation while we jointly eliminate source samples which are unrelated to target samples by estimating those weights. By introducing regularization which induces the weights to be homogeneous,...
We solve a manifold learning problem by searching for hypersurfaces fitted to the data. The method, called support vector manifold learning (SVML), transforms data to a kernel-induced feature space, duplicates points, shifts them in two opposite directions and solves a classification problem using support vector machines (SVM). Then, we cluster data by mapping found hypersurfaces to clusters, the...
We study tiled algorithms for going from a "full" matrix to a condensed "band bidiagonal" form using orthog-onal transformations: (i) the tiled bidiagonalization algorithm BIDIAG, which is a tiled version of the standard scalar bidiago-nalization algorithm; and (ii) the R-bidiagonalization algorithm R-BIDIAG, which is a tiled version of the algorithm which consists in first performing...
Applications of pattern recognition, design of faulttolerant systems and communications have key problems that arenaturally described by partially defined (incompletely defined)discrete functions. Such partially defined functions arising frompractical demands usually have a large number of variables andso their direct implementations require complex systems. Thusit is important to have at hand an...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.