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.
For i-vector model, normalization approach is Probabilistic linear discriminant analysis and has a significant performance for verification of speaker. However it requires a huge development data which cost a lot in many cases. Unsupervised adaption method is a possible approach, which use unlabeled data to adapt PLDA scattering matrices to the target domain. In this paper, ‘local training’ approach...
Mathematical modeling is an important approach for creating a parallel database management system that could efficiently use capabilities, provided by modern heterogeneous computational clusters, equipped with manycore coprocessors or GPUs. To this day, several models heterogeneous computational systems were proposed, but none of them is suited for modeling database processing. In this paper, we address...
A lot of research has been dedicated to finding an optimal strategy to defend network infrastructure. The proposed methods are usually evaluated using simulations, replayed attacks or testbed environments. However, these evaluation methods may give biased results, because in real life, attackers can follow a suboptimal strategy or react to a defence in an unexpected way. In this paper, we use a network...
Due to the number of cloud providers, as well as the extensive collection of services, cloud computing provides very flexible environments, where resources and services can be provisioned and released on demand. However, reconfiguration and adaptation mechanisms in cloud environments are very heterogeneous and often exhibit complex constraints. For example, when reconfiguring a cloud system, a set...
Model-based approaches to Speaker Verification (SV), such as Joint Factor Analysis (JFA), i-vector and relevance Maximum-a-Posteriori (MAP), have shown to provide state-of-the-art performance for text-dependent systems with fixed phrases. The performance of i-vector and JFA models has been further enhanced by estimating posteriors from Deep Neural Network (DNN) instead of Gaussian Mixture Model (GMM)...
Initialization and feature selection are crucial in supervised landmark detection. Mismatching caused by detector error and discrepant initialization is very common in these existing methods. To solve this problem, we proposed a new method, which includes a new initialize model and multitask feature learning, for the robust facial landmark localization. In our new method, firstly, a fast detection...
With the progressive development of information and communication technologies, we are now forming a new world called hyperworld that is composed by the cyber world and the physical world with various digital explosions including data, connectivity, service and intelligence. Therefore, Cyber-I has been proposed, which is a real individual's counterpart in cyberspace, and is to create a unique, digital,...
Predicting memory occupancy during the execution of large-scale analytical workloads becomes critical for in-memory databases. In particular, probabilistic performance measures for such systems are of interest, but difficult to model with analytical methods due to the highly variable threading levels in corresponding workloads. Since literature with queueing theoretic background largely ignores the...
This paper presents local structure tensor (LST) based methods for extracting the anisotropy and orientation information characterizing a textured image sample. A Log-Euclidean (LE) multivariate Gaussian model is proposed for representing the marginal distribution of the LST field of a texture. An extended model is considered as well for describing its spatial dependencies. The potential of these...
We propose an efficient adaptive oblivious transfer protocol with hidden access policies. This scheme allows a receiver to anonymously recover a message from a database which is protected by hidden attribute based access policy if the receiver's attribute set satisfies the associated access policy implicitly. The proposed scheme is secure in the presence of malicious adversary under the q-Strong Diffie-Hellman...
Datamining (DM) brings together a wide range of techniques and algorithms which allow the extraction of knowledge from databases for timely decision making. DM has been applied to different fields of study. One important research field is Education. Applying DM in education is known as educational datamining (EDM). The main purpose of EDM is to analyze data from educational institutions using different...
Gesticulation, together with the speech, is an important part of natural and affective human-human interaction. Analysis of gesticulation and speech is expected to help designing more natural human-computer interaction (HCI) systems. We build the JestKOD database, which consists of speech and motion capture recordings of dyadic interactions. In this paper we describe our annotation efforts and present...
In this paper a speaker dependent cohort selection for T-norm score normalization is proposed in the context of text-dependent speaker verification. The goal of the proposed technique is to find a set of cohort speakers who are close to the target speaker. In order to properly select the subset of speakers for the normalization, a distance between each target speaker model and the the available normalization...
In this paper, fast ellipse fitting based pupil tracking design is studied for human-computer interaction applications. The proposed scheme uses the two-stage Random Sample Consensus (RANSAC) based fast ellipse fittings. At the first stage, the binaried pupil contour based ellipse model is applied to select the possible candidate points on Canny filtered edges. At the second stage, the Canny edge...
The current operation and development of National Defense M&S (Modeling & Simulation) virtual test application runs by FOM (Federation Object Model) type through HLA/RTI during simulation to interchange data. Simulation data interchanged by this method is one-shot data and won't be stored persistently. However, FOM typed data is in need of analysis, processing and reusability for the purpose...
This paper examines the assembly of interdisciplinary teams in emerging scientific fields. We develop and validate a hybrid systems dynamics and agent-based computational model using data over a 15 year period from the assembly of teams in the emerging scientific field of Oncofertility. We found that, when a new field emerges, team assembly is influenced by the reputation and seniority of the researchers,...
This paper aims to show that regularities emerge from the strongly stochastic properties of human behavior. To this aim, the paper presents the application of a new Knowledge Discovery in Database (KDD) process, called Timed Observations Mining for Learning (TOM4L), on the timed data provided by a smart building of offices of the southeast of France during 12 months from April 2011 to March 2012....
Visual-based 3D model retrieval presents great potential, for its broad application prospects and relatively high accuracy. A branch of visual-based methods utilizes scale invariant feature transform (SIFT) on 2D rendered images of a 3D model viewed from regularly sampled locations on a sphere, and then the bag-of-words framework is employed to improve the retrieval precision. However, in existing...
We present a transaction model which simultaneously supports different consistency levels, which include serial-izable transactions for strong consistency, and weaker consistency models such as causal snapshot isolation (CSI), CSI with commutative updates, and CSI with asynchronous updates. This model is useful in managing large-scale replicated data with different consistency guarantees to make suitable...
Information over the web is increasingly retrieved from relational databases in which the query language is based on exact matching, data fulfil completely the query or not. The results returned to the user contain only tuples that satisfy the conditions of the query. Thereby, the user can be confronted to the problem of empty answers in the case of too selective query. To overcome this problem, several...
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.