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.
People detection in single 2D images has improved greatly in recent years. However, comparatively little of this progress has percolated into multi-camera multipeople tracking algorithms, whose performance still degrades severely when scenes become very crowded. In this work, we introduce a new architecture that combines Convolutional Neural Nets and Conditional Random Fields to explicitly model those...
With the development of neural networks based machine learning and their usage in mission critical applications, voices are rising against the black box aspect of neural networks as it becomes crucial to understand their limits and capabilities. With the rise of neuromorphic hardware, it is even more critical to understand how a neural network, as a distributed system, tolerates the failures of its...
In this paper the calculations of the robustness of a network is addressed. After a brief description of the most relevant metrics, our Network Robustness Simulator(NRS) is presented as well as its structure and working model. The NRS computes the robustness I a dynamic scenario, it copes with multiple failures and different types of attack. In particular, the addition of the epidemic based model...
This work presents a dead-time compensation technique for robust predictive current control of grid-connected voltage source inverter. Among current controls, dead-beat predictive controllers are one of the fastest, but are extremely sensitive to inconsistencies between the model and the actual plant. Dead-time modifies the plant model and is one of the main sources of distortion in high dynamic range...
Today, Internet of Things (IoT) is an emergent concept in which billions of devices are connected to Internet capable of producing and exchanging data. One of the most used technologies in this area regards to the Radio Frequency Identification (RFID). It can produce large amount of data from many things like objects, persons and assets. Thus, it is needed middlewares which must support processing...
The Cerebellar Model Articulation Controller (CMAC) is a type of neural network particularly suited to real-time control applications due to fast adaptation and the ability to handle many inputs. However, the CMAC is well-known to exhibit weight (adaptive-parameter) drift when used in adaptive control, and overlearning when applied in static learning situations. A weight smoothing algorithm originally...
The paper discusses a novel probabilistic approach for online parameter estimation of the predictor model used in an MPC (Model Predictive Control) setting in the presence of model uncertainties and external disturbances. Model uncertainty makes it hard to compute an optimal control in general case, because it is needed to take into account all possible values of model parameters. Therefore, it is...
Previous models based on Deep Convolutional Neural Networks (DCNN) for face verification focused on learning face representations. The face features extracted from the models are applied to additional metric learning to improve a verification accuracy. The models extract high-dimensional face features to solve a multi-class classification. This results in a dependency of a model on specific training...
Semi-symbolic simulation is becoming popular for inclusion of parameter uncertainties in the system design analysis. For robust control system design optimization, computational methods enabling fast semi-symbolic simulations are necessary. We propose an operational computation method based on orthogonal signals that is faster than step integration methods and allows the direct evaluation of system...
This paper proposes a systematic approach to derive analytical and explicit certificates for transient stability analysis of inverter-based microgrids. We first derive a line-based model of microgrids with general mesh structure. Then by employing Lyapunov stability theory with a quadratic Lyapunov function, the region of attraction of any post-fault stable equilibrium point (EP) is estimated by a...
Gaussian Processes (GPs) are state-of-the-art tools for regression. Inference of GP hyperparameters is typically done by maximizing the marginal log-likelihood (ML). If the data truly follows the GP model, using the ML approach is optimal and computationally efficient. Unfortunately very often this is not case and suboptimal results are obtained in terms of prediction error. Alternative procedures...
Speaker diarization systems aim to segment an audio signal into homogeneous sections with only one active speaker and answer the question "who spoke when?" We present a novel approach to speaker diarization exploiting spatial information through robust statistical modeling of Time Difference of Arrival (TDOA) estimates obtained using pairs of microphones. The TDOAs are modeled with Gaussian...
Digital predistortion (DPD) is an effective power amplifier (PA) linearization technique improving the system energy efficiency. At this point, real-time DPD adaptation is still an open issue due to the high computational complexity during the coefficients estimation procedure. Online censoring approach, which is effective in reducing the redundant data samples, can be applied in the DPD coefficients...
Radial Basis Function(RBF) mesh deformation method has been widely used in CFD simulations with moving boundaries due to its high robustness and accuracy. The original implementation of the RBF mesh deformation method in OpenFOAM(a widely used CFD software) is purely serial with relatively low computational performance. To reduce the time cost of the mesh motion in large-scale simulations, this paper...
The interpolation of correspondences (EpicFlow) was widely used for optical flow estimation in most-recent works. It has the advantage of edge-preserving and efficiency. However, it is vulnerable to input matching noise, which is inevitable in modern matching techniques. In this paper, we present a Robust Interpolation method of Correspondences (called RicFlow) to overcome the weakness. First, the...
Images are formed by counting how many photons traveling from a given set of directions hit an image sensor during a given time interval. When photons are few and far in between, the concept of image breaks down and it is best to consider directly the flow of photons. Computer vision in this regime, which we call scotopic, is radically different from the classical image-based paradigm in that visual...
We propose a novel and principled hybrid CNN+CRF model for stereo estimation. Our model allows to exploit the advantages of both, convolutional neural networks (CNNs) and conditional random fields (CRFs) in an unified approach. The CNNs compute expressive features for matching and distinctive color edges, which in turn are used to compute the unary and binary costs of the CRF. For inference, we apply...
Most of the recent successful methods in accurate object detection and localization used some variants of R-CNN style two stage Convolutional Neural Networks (CNN) where plausible regions were proposed in the first stage then followed by a second stage for decision refinement. Despite the simplicity of training and the efficiency in deployment, the single stage detection methods have not been as competitive...
In this work, we introduce a highly efficient algorithm to address the nonnegative matrix underapproximation (NMU) problem, i.e., nonnegative matrix factorization (NMF) with an additional underapproximation constraint. NMU results are interesting as, compared to traditional NMF, they present additional sparsity and part-based behavior, explaining unique data features. To show these features in practice,...
The CNN-encoding of features from entire videos for the representation of human actions has rarely been addressed. Instead, CNN work has focused on approaches to fuse spatial and temporal networks, but these were typically limited to processing shorter sequences. We present a new video representation, called temporal linear encoding (TLE) and embedded inside of CNNs as a new layer, which captures...
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.