Sharif Digital Repository

Digital video stabilization is a category of techniques used to reduce the impact of unintentional camera motion such as jitter, jiggle, and other unsteady motions. These unintentional shakings degrade visual quality of videos and reduce the performance of subsequent processes such as video compression. Digital video stabilization which is performed by post processing the acquired frames, suffers from inaccuracy of motion estimation which is mostly due to the local motions of internal moving objects included in videos, and long processing time which prohibits them from being used in real time applications. In this paper we propose a fast and accurate transform based motion estimation method... 

A Bayesian framework for 3-D human pose estimation from monocular images based on sparse representation (SR) is introduced. Our probabilistic approach aims at simultaneously learning two overcomplete dictionaries (one for the visual input space and the other for the pose space) with a shared sparse representation. Existing SR-based pose estimation approaches only offer a point estimation of the dictionary and the sparse codes. Therefore, they might be unreliable when the number of training examples is small. Our Bayesian framework estimates a posterior distribution for the sparse codes and the dictionaries from labeled training data. Hence, it is robust to overfitting on small-size training... 

This paper proposes a global motion estimation method to remove unintentional camera motions which degrade the visual quality of image sequences. The proposed approach is based on combination of 2D Radon transform, 1D Fourier transform and 1D Scale transform which can accurately estimate scale, rotational and translational distortions of camera motion and is robust to internal moving objects. Our experimental results with real and synthesized videos indicate the effectiveness of our proposed method  

While recent supervised dictionary learning methods have attained promising results on the classification tasks, their performance depends on the availability of the large labeled datasets. However, in many real world applications, accessing to sufficient labeled data may be expensive and/or time consuming, but its relatively easy to acquire a large amount of unlabeled data. In this paper, we propose a probabilistic framework for discriminative dictionary learning which uses both the labeled and unlabeled data. Experimental results demonstrate that the performance of the proposed method is significantly better than the state of the art dictionary based classification methods  

Distance metric learning is a powerful approach to deal with the clustering problem with side information. For semi-supervised clustering, usually a set of pairwise similarity and dissimilarity constraints is provided as supervisory information. Although some of the existing methods can use both equivalence (similarity) and inequivalence (dissimilarity) constraints, they are usually limited to learning a global Mahalanobis metric (i.e., finding a linear transformation). Moreover, they find metrics only according to the data points appearing in constraints, and cannot utilize information of other data points. In this paper, we propose a probabilistic metric learning algorithm which uses... 

In many real-world applications (e.g. social media application), data usually consists of diverse input modalities that originates from various heterogeneous sources. Learning a similarity measure for such data is of great importance for vast number of applications such as classification, clustering, retrieval, etc. Defining an appropriate distance metric between data points with multiple modalities is a key challenge that has a great impact on the performance of many multimedia applications. Existing approaches for multi-modal distance metric learning only offer point estimation of the distance matrix and/or latent features, and can therefore be unreliable when the number of training... 

In this paper, we present a new method to solve tracking problem. The proposed method combines sparse representation and motion estimation to track an object. Recently. sparse representation has gained much attention in signal processing and computer vision. Sparse representation can be used as a classifier but has high time complexity. Here, we utilize motion information in order to reduce this computation time by not calculating sparse codes for all the frames. Experimental results demonstrates that the achieved result are accurate enough and have much less computation time than using just a sparse classifier  

γ-Alumina is used to catalyze the Cannizzaro reaction in the absence of any base under microwave irradiation in high yields. In the case of terephthalaldehyde the reaction is carried out with high selectivity  

Enolizable ketones have been reacted in a one-pot method with aromatic aldehydes, acetyl chloride and acetonitrile at room temperature in the presence of SnCl4/SiO2 to furnish the corresponding β-acetamidoketones in improved yields. Acetylation of an aromatic hydroxyl group was observed while using 4-hydroxybenzaldehyde or vanillin and the corresponding β-acetamidoketones were isolated in an excellent yield  

The boundary layer integral method is used to investigate the development of the turbulent swirling flow at the entrance region of a conical nozzle. The governing equations in the spherical coordinate system are simplified with the boundary layer assumptions and integrated through the boundary layer. The resulting sets of differential equations are then solved by the fourth-order Adams predictor-corrector method. The free vortex and uniform velocity profiles are applied for the tangential and axial velocities at the inlet region, respectively. Due to the lack of experimental data for swirling flows in converging nozzles, the developed model is validated against the numerical simulations. The... 

The governing equations of two-dimensional steady density currents are solved numerically using a finite volume method. The v2̄-f turbulence model, based on standard k - s model, is used for the turbulence closure. In this method, all Reynolds stress equations are replaced with both a transport equation for v2̄ and an elliptic relaxation equation for f, a parameter closely related to the pressure strain redistribution term. The Simple-C procedure is used for pressure-velocity coupling. In addition, Boussinesq's approximation is used to obtain the momentum equation. The computed height of the progressive density current is compared to the measured data in the literature, resulting in good... 

The deposition behavior of fine sediment is an important phenomenon, and yet unclear to engineers concerned about reservoir sedimentation. An elliptic relaxation turbulence model (v2̄ - f model) has been used to simulate the motion of turbid density currents laden whit fine solid particles. During the last few years, the v2̄ - f turbulence model has become increasingly popular due to its ability to account for near-wall damping without use of damping functions. In addition, it has been proved that the v2̄ - f model to be superior to other RANS methods in many fluid flows where complex flow features are present. Due to low Reynolds number turbulence of turbidity current,(its critical Reynolds... 

In this paper, an analytical solution for steady creeping motion of viscoelastic drop falling through a viscous Newtonian fluid is presented. The Oldroyd-B model is used as the constitutive equation. The analytical solutions for both interior and exterior flows are obtained using the perturbation method. Deborah number and capillary numbers are considered as the perturbation parameters. The effect of viscoelastic properties on drop shape and motion are studied in detail. The previous empirical studies indicated that unlike the Newtonian creeping drop in which the drop shape is exactly spherical, a dimpled shape appears in viscoelastic drops. It is shown that the results of the present... 

In this paper, creeping motion of a viscoelastic drop falling through a Newtonian fluid is investigated experimentally and analytically. A polymeric solution of 0.08 % xanthan gum in 80:20 glycerol/water and silicon oil is implemented as the viscoelastic drop and the bulk viscous fluids, respectively. The shape and motion of falling drops are visualized using a high speed camera. The perturbation technique is employed for both interior and exterior flows, and Deborah and capillary numbers are considered as perturbation parameters up to second order. The product of Deborah and capillary numbers is also used as a perturbation parameter to apply the boundary condition on the deformation on the... 

In this paper, we consider the problem of minimizing the uplink delays of users in a 5G cellular network. Such cellular network is based on a Cloud Radio Access Network (CRAN) architecture with limited fronthaul capacity, where our goal is to minimize delays of all users through an optimal resource allocation. Earlier works minimize average delay of each user assuming same transmit power for all users. Combining Pareto optimization and Markov Decision Process (MDP), we show that every desired balance in the trade-off among infinite-horizon average-reward delays, is achievable by minimizing a properly weighted sum delays. In addition, we solve the problem in two realistic scenarios;... 

Microwave induced Cannizzaro reaction, which was performed in presence of neutral γ-alumina as a polymeric catalyst, was discussed. Cannizzaro reaction was carried out on alumina surface and other unknown products were also obtained. Solvent extraction method was utilized to isolate mild and selective benzyl alcohol and carboxylic acids  

A well known method to reduce the intrinsic complexity of Multiple Input Multiple Output (MIMO) systems is to choose a subset of available antennas which have stronger links than the others, in order to perform the specified MIMO algorithm. The data resulted from the antenna selection process (at the receiver side) is sent back to the transmitter side via a feedback channel. There seems to be a need to reduce the number of feedack bits, specially when the number of antennas is not small. In this paper, we investigate the problem of reducing the number of feedback bits in antenna selection techniques. We've proposed two methods using vector quantization techniques to perform feedback bit... 

Nowadays, deployment of peer-to-peer video streaming systems over wireless mesh networks has attained raising popularity among large number of users around the world. In this paper, we present an efficient peer-to-peer live video streaming architecture over multi-hop wireless mesh networks. In our proposed architecture, we take the physical topology of network into account and based on a distributed distributed locality-aware neighbor selection algorithm in the overlay construction phase, we generate an efficient mesh-based overlay on top of wireless mesh networks. In locality-aware neighbor selection algorithm, instead of choosing randomly, peers find their best neighbors based on their... 

Standard sparse decomposition (with applications in many different areas including compressive sampling) amounts to finding the minimum ℓ 0-norm solution of an underdetermined system of linear equations. In this decomposition, all atoms are treated 'uniformly' for being included or not in the decomposition. However, one may wish to weigh more or less certain atoms, or, assign higher costs to some other atoms to be included in the decomposition. This can happen for example when there is prior information available on each atom. This motivates generalizing the notion of minimal ℓ 0-norm solution to that of minimal weighted ℓ 0-norm solution. On the other hand, relaxing weighted ℓ 0-norm via...