Facial attributes can provide rich ancillary information which can be utilized for different applications such as targeted marketing, human computer interaction, and law enforcement. This research focuses on facial attribute prediction using a novel deep learning formulation, termed as R-Codean autoencoder. The paper first presents Cosine similarity based loss function in an autoencoder which i...

Residuals are frequently used to evaluate the validity of the assumptions of statistical models and may also be employed as tools for model selection. For standard (normal) linear models, for example, residuals are used to verify homoscedasticity, linearity of effects, presence of outliers, normality and independence of the errors. Similar uses may be envisaged for three types of residuals that...

The paper describes a residual analysis-synthesis system which exploits the human perception mechanism on temporal varying signals using Zwicker's three dimensional excitation-critical-bandrate time pattern as the framework. Temporal information is retrieved using a linear predictive analysis on the discrete cosine transformed signal and critical band intensity information is obtained by using ...

Control by eddy current is one of techniques that’s very used in nondestructive evaluation of materials .These processes find an application significant in the determination of the mechanical and metallurgical parameters of materials. The object of this work is to carry out an analysis of the residual stresses in the material created by external solicitations and analysed by eddy currents as a ...

We analyze the kinematic properties of offset-domain Common Image Gathers (CIGs) and Angle-Domain CIGs (ADCIGs) computed by wavefield-continuation migration. Our results are valid regardless of whether the CIGs were obtained by using the correct migration velocity. They thus can be used as a theoretical basis for developing Migration Velocity Analysis (MVA) methods that exploit the velocity inf...

A Rasch Poisson counts (RPC) model is described to identify individual latent traits and facilities of the items tests that error (or success) count in several tasks over time, instead modeling correct responses a test as dichotomous item response theory (IRT) model. These types can be more informative than traditional tests. To estimate parameters, we consider Bayesian approach using integrate...