Sharif Digital Repository

Multiple human 3D pose estimation is a challenging task. It is mainly because of large variations in the scale and pose of humans, fast motions, multiple persons in the scene, and arbitrary number of visible body parts due to occlusion or truncation. Some of these ambiguities can be resolved by using multiview images. This is due to the fact that more evidences of body parts would be available in multiple views. In this work, a novel method for multiple human 3D pose estimation using evidences in multiview images is proposed. The proposed method utilizes a fully connected pairwise conditional random field that contains two types of pairwise terms. The first pairwise term encodes the spatial... 

Although the read disturb spin-transfer torque RAM approximation technique improves performance, it may consist of an approximate read plus an approximate write both at the same time. So it may degrade the application quality of result (QoR) considerably. On the other hand, the incorrect read decision approximation technique improves power without corrupting the stored data. We adopt an opportunity study for instruction-based distinction of read implementation to take advantage of both of the approximation techniques, while enhancing application's QoR. We evaluated the proposed method using a set of state-of-the-art benchmarks. The experimental results show that our method allows to increase... 

One of the most challenging tasks in computer vision is human action recognition. The recent development of depth sensors has created new opportunities in this field of research. In this paper, a novel supervised spatio-temporal kernel descriptor (SSTKDes) is proposed from RGB-depth videos to establish a discriminative and compact feature representation of actions. To enhance the descriptive and discriminative ability of the descriptor, extracted primary kernel-based features are transformed into a new space by exploiting a supervised training strategy; i.e., large margin nearest neighbor (LMNN). The LMNN highly reduces the error of a nearest neighbor classifier by minimizing the intra-class... 

In cryptographic engineering, extensive attention has been devoted to ameliorating the performance and security of the algorithms within. Nonetheless, in the state-of-the-art, the approaches for increasing the reliability of the efficient hash functions ECHO and Fugue have not been presented to date.We propose efficient fault detection schemes by presenting closed formulations for the predicted signatures of different transformations in these algorithms. These signatures are derived to achieve low overhead for the specific transformations and can be tailored to include byte/word-wide predicted signatures. Through simulations, we show that the proposed fault detection schemes are... 

Big-data server applications frequently encounter data misses, and hence, lose significant performance potential. One way to reduce the number of data misses or their effect is data prefetching. As data accesses have high temporal correlations, temporal prefetching techniques are promising for them. While state-of-the-art temporal prefetching techniques are effective at reducing the number of data misses, we observe that there is a significant gap between what they offer and the opportunity. This work aims to improve the effectiveness of temporal prefetching techniques. We identify the lookup mechanism of existing temporal prefetchers responsible for the large gap between what they offer and... 

In this paper, we propose an algorithm for recovery of sparse and low-rank components of matrices using an iterative method with adaptive thresholding. In each iteration of the algorithm, the low-rank and sparse components are obtained using a thresholding operator. The proposed algorithm is fast and can be implemented easily. We compare it with the state-of-the-art algorithms. We also apply it to some applications such as background modeling in video sequences, removing shadows and specularities from face images, and image restoration. The simulation results show that the proposed algorithm has a suitable performance with low run-time. © 2017 Elsevier Inc  

Multi-label classification is a learning task in which each data sample can belong to more than one class. Until now, some methods that are based on reducing the dimensionality of the label space have been proposed. However, these methods have not used specific properties of the label space for this purpose. In this paper, we intend to find a hidden space in which both the input feature vectors and the label vectors are embedded. We propose a modified Non-Negative Matrix Factorization (NMF) method that is suitable for decomposing the label matrix and finding a proper hidden space by a feature-aware approach. We consider that the label matrix is binary and also in this matrix some deserving... 

The approximation power of general feedforward neural networks with piecewise linear activation functions is investigated. First, lower bounds on the size of a network are established in terms of the approximation error and network depth and width. These bounds improve upon state- of-the-art bounds for certain classes of functions, such as strongly convex functions. Second, an upper bound is established on the difference of two neural networks with identical weights but different activation functions. © The Author(s) 2018  

Due to the unprecedented speed and volume of generated raw data in most of applications, data stream mining has attracted a lot of attention recently. Methods for solving these problems should address challenges in this area such as infinite length, concept-drift, recurring concepts, and concept-evolution. Moreover, due to the speedy intrinsic of data streams, the time and space complexity of the methods are extremely important. This paper proposes a novel method based on multiple-kernels for classifying non-stationary data streams, which addresses the mentioned challenges with special attention to the space complexity. By learning multiple kernels and specifying the boundaries of classes in... 

Reconstructing a signal corrupted by impulsive noise is of high importance in several applications, including impulsive noise removal from images, audios and videos, and separating texts from images. Investigating this problem, in this paper we propose a new method to reconstruct a noise-corrupted signal where both signal and noise are sparse but in different domains. We apply our algorithm for impulsive noise (Salt- and-Pepper Noise (SPN) and Random-Valued Impulsive Noise (RVIN) removal from images and compare our results with other notable algorithms in the literature. Simulation indicates show that our algorithm is not only simple and fast, but also it outperforms the other... 

This paper presents a simple yet efficient method for an anomaly-based Intrusion Detection System (IDS). In reality, IDSs can be defined as a one-class classification system, where the normal traffic is the target class. The high diversity of network attacks in addition to the need for generalization, motivate us to propose a semi-supervised method. Inspired by the successes of Generative Adversarial Networks (GANs) for training deep models in semi-unsupervised setting, we have proposed an end-to-end deep architecture for IDS. The proposed architecture is composed of two deep networks, each of which trained by competing with each other to understand the underlying concept of the normal... 

Today's world energy-related challenges, ranging from anthropogenic climate change to continuous growth of demand for different energy forms, have enforced planners of energy systems (ESs) to concentrate on more optimal and eco-friendly operation and/or expansion planning methodologies. In this context, increased interdependencies of gas, heat and electricity ESs have recently encouraged the planners to design operation and/or expiation strategies in an integrated manner in favor of a new concept, the so-called “Energy Hub” (EH). Although this concept has been employed so far in a multitude of studies, but its real nature, advantages, difficulties, importance or inevitability aspect, and... 

Differentiating between real and fake news propagation through online social networks is an important issue in many applications. The time gap between the news release time and detection of its label is a significant step towards broadcasting the real information and avoiding the fake. Therefore, one of the challenging tasks in this area is to identify fake and real news in early stages of propagation. However, there is a tradeoff between minimizing the time gap and maximizing accuracy. Despite recent efforts in detection of fake news, there has been no significant work that explicitly incorporates early detection in its model. The proposed method utilizes recurrent neural networks with a... 

The effective utilization of the spectrum has become an essential goal in the communications field, which is addressed by the Cognitive Radio (CR) systems. The primary task in a CR system is to sense the spectrum to identify its holes to be exploited by the secondary users. In this paper, we tackle the compressed spectrum sensing problem in a cooperative manner. The CRs distributed in an area take the samples of the signal that has been reached to them through a wireless fading channel. The spectrum has the block-sparse structure. Moreover, the spectrum observed by different CRs in an area share the same block-sparse support. Therefore, we suggest to exploit the joint block-sparsity... 

In this paper, we propose two methods for signal denoising where both signal and noise are sparse but in different domains. First, an optimization problem is proposed which is non-convex and NP-hard due to the existence of ℓ 0 norm in its cost function. Then, we propose two approaches to approximate and solve it. We also provide the proof of convergence for the proposed methods. The problem addressed in this paper arises in some applications for example in image denoising where the noise is sparse, signal reconstruction in the case of random sampling where the random mask is unknown, and error detection and error correction in the case of missing samples. The experimental results indicate... 

Multi-modal data comprising imaging (MRI, fMRI, PET, etc.) and non-imaging (clinical test, demographics, etc.) data can be collected together and used for disease prediction. Such diverse data gives complementary information about the patient's condition to make an informed diagnosis. A model capable of leveraging the individuality of each multi-modal data is required for better disease prediction. We propose a graph convolution based deep model which takes into account the distinctiveness of each element of the multi-modal data. We incorporate a novel self-attention layer, which weights every element of the demographic data by exploring its relation to the underlying disease. We demonstrate... 

Applications extensively use data objects with a regular and fixed layout, which leads to the recurrence of access patterns over memory regions. Spatial data prefetching techniques exploit this phenomenon to prefetch future memory references and hide the long latency of DRAM accesses. While state-of-the-art spatial data prefetchers are effective at reducing the number of data misses, we observe that there is still significant room for improvement. To select an access pattern for prefetching, existing spatial prefetchers associate observed access patterns to either a short event with a high probability of recurrence or a long event with a low probability of recurrence. Consequently, the... 

We present our work on improving the numerator graph for discriminative training using the lattice-free maximum mutual information (MMI) criterion. Specifically, we propose a scheme for creating unconstrained numerator graphs by removing time constraints from the baseline numerator graphs. This leads to much smaller graphs and therefore faster preparation of training supervisions. By testing the proposed un-constrained supervisions using factorized time-delay neural network (TDNN) models, we observe 0.5% to 2.6% relative improvement over the state-of-the-art word error rates on various large-vocabulary speech recognition databases. © 2018 IEEE  

Recently, energy harvesting systems that utilize hybrid NVM-SRAM memory in their designs are introduced as a promising alternative for battery-operated systems. Since the ambient input power of an energy harvesting system fluctuates as the environmental conditions change, the system may stop the execution of programs until it receives enough energy to continue the execution. Resuming the execution of a program after the suspension may lead to data inconsistency in an energy harvesting system and threatens the correct functionality of the programs. In this paper, we propose COACH, an energy-efficient consistency-aware memory scheme which guarantees the correct functionality and consistency of... 

In this study, the optimal sensor placement problem for multi-source angle of arrival localisation is investigated. The authors adopt the A-optimality criterion, maximising the trace of Fisher information matrix, to determine the optimal sensor-target geometry under distance-dependent Gaussian noise model. A recursive representation of the Cramer-Rao lower bound is derived to recast the sensor placement problem into a sequential method, obtaining the optimal sensor geometries in a step by step manner. Note that the state-of-the-art methods are highly sensitive to the source location changes such that they should be relocated by any later changes in target geometries, which is practically...