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.
Convolutional Neural Network (CNN) has led to significant progress in face recognition. Currently most CNNbased face recognition methods follow a two-step pipeline, i.e. a detected face is first aligned to a canonical one predefined by a mean face shape, and then it is fed into a CNN to extract features for recognition. The alignment step transforms all faces to the same shape, which can cause loss...
Apparent age estimation from face image has attracted more and more attentions as it is favorable in some real-world applications. In this work, we propose an end-to-end learning approach for robust apparent age estimation, named by us AgeNet. Specifically, we address the apparent age estimation problem by fusing two kinds of models, i.e., real-value based regression models and Gaussian label distribution...
This paper presents a novel Current-Mirror (CM) based Master-Slave Level Conversion Flip Flop (MS-LCFF) to perform data latching and level shifting from sub-threshold voltage, to near-threshold voltage, and up to super-threshold voltage. The CM-based MS-LCFF enables energy-efficient ultra-low-voltage operation by applying dual-supply and multi-supply designs into sub/near-threshold regions. Simulation...
In this paper, a novel foreground detection method based on two-stage framework is presented. In the first stage, a class of structured sparsity-inducing norms is introduced to model moving objects in videos and thus regard the observed sequence as being made up of the sum of a low-rank matrix and a structured sparse outlier matrix. In virtue of adaptive parameters, the proposed method includes a...
In this study, an LMI (Linear Matrix Inequality) based H∞ robust controller design approach is proposed to improve the performance of the designed U-pole placement control systems. Unlike the classical design procedures, the control-oriented U-model based nonlinear control systems cancel the nonlinearity of the nonlinear models. Therefore, the closed loop transfer function of U-pole placement control...
In this study, an initial robust stability analysis procedure is proposed to test the performance of the designed U-pole placement control systems. Unlike the classical design procedures for non-linear control systems, the control-oriented U-model based non-linear control systems cancel the non-linearity of the non-linear models. Therefore, the closed loop transfer function of U-pole placement control...
Lip tracking is crucial to the success of a lipreading recognition system. This paper presents a robust lip tracking algorithm using localized color active contours and deformable models. The proposed approach utilizes a combined semi-ellipse as the initial evolving curve and computes the localized energies in color space for evolving such that a separation of the original lip image into lip and non-lip...
Open-loop unstable systems with time delay are difficult to control using routine control methods. To solve this problem, an improved internal model control scheme, which adds a PD controller to traditional IMC structure, is proposed. Set point tracking performance and robustness can be designed independently. The tuning formula of PD controller is confirmed by a new performance index function. Fuzzy...
In this paper, a novel robust audio watermarking algorithm based on discrete cosine transform (DCT) and vector quantization (VQ) is proposed. The original audio signal is first segmented into frames. The embedding process for each frame is as follows. First, the original frame is DCT transformed, and several middle-frequency coefficients are composed of a vector to be modified. Then the labeled-codeword...
In this paper, a novel robust audio watermarking algorithm is presented. It takes full advantage of the multi-resolution and the energy compression properties of the discrete wavelet transform (DWT) and discrete cosine transform (DCT) respectively, and embeds a pseudorandom permuted binary image in the original audio signal by vector quantization (VQ). The original audio is first transformed using...
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.