ALBERT

Description: We examine a distributed detection problem in a wireless sensor network, where sensor nodes collaborate to detect a Gaussian signal with an unknown change of power, i.e., a scale parameter. Due to power/bandwidth constraints, we consider the case where each sensor quantizes its observation into a binary digit. The binary data are then transmitted through error-prone wireless links to a fusion center, where a generalized likelihood ratio test (GLRT) detector is employed to perform a global decision. We study the design of a binary quantizer based on an asymptotic analysis of the GLRT. Interestingly, the quantization threshold of the quantizer is independent of the unknown scale parameter. Numerical results are included to illustrate the performance of the proposed quantizer and GLRT in binary symmetric channels (BSCs).

Description: More and more hybrid electric vehicles are driven since they offer such advantages as energy savings and better active safety performance. Hybrid vehicles have two or more power driving systems and frequently switch working condition, so controlling stability is very important. In this work, a two-stage Kalman algorithm method is used to fuse data in hybrid vehicle stability testing. First, the RT3102 navigation system and Dewetron system are introduced. Second, a modeling of data fusion is proposed based on the Kalman filter. Then, this modeling is simulated and tested on a sample vehicle, using Carsim and Simulink software to test the results. The results showed the merits of this modeling.

Description: In this paper, a comprehensive comparison analysis in terms of outage probability and average symbol error ratio (SER) is presented for cooperative cognitive multiple-input and multiple-output (CC-MIMO) multiuser systems with amplify-and-forward (AF) protocol. Specially, we consider two scenarios where the CC-MIMO multiuser systems have the perfect and imperfect channel state information (CSI). The CC-MIMO multiuser systems consist of one multi-antenna source, one single-antenna relay, and multiple multi-antenna destinations. At the secondary source and destinations, the maximal ratio transmission (MRT) and maximal ratio combining (MRC) are employed, respectively. For such CC-MIMO multiuser systems, we first obtain the exact closed-form expressions of outage probability under the two cases where the CC-MIMO multiuser systems have the perfect and imperfect CSI. Then, to reduce the implementation complexity, the tight lower bounds of outage probability and average SER are derived. Finally, to obtain insight, by using the high signal-to-noise ratio (SNR) approximation, the asymptotic estimations of outage probability are achieved. The numerical results show that the derivations are agreed with the simulations, which validate our derivations. At the same time, the results show that, for the systems without perfect CSI, the achievable diversity order reduces to one, regardless of the number of antennas at the cognitive source and destinations as well as the number of the cognitive destinations. Nevertheless, these key parameters affect the coding gain of the CC-MIMO multiuser systems. When the systems have the perfect CSI (or without feedback delay), the achievable diversity gain is determined by the minimum between the number of source’s antennas and the product of the number of destinations and the number of destination’s antennas. For the effect of PU’s parameters, our results indicate that primary systems only affect the coding gain but not the diversity gain.

Description: The goal of cross-domain matching (CDM) is to find correspondences between two sets of objects in different domains in an unsupervised way. CDM has various interesting applications, including photo album summarization where photos are automatically aligned into a designed frame expressed in the Cartesian coordinate system, and temporal alignment which aligns sequences such as videos that are potentially expressed using different features. In this paper, we propose an information-theoretic CDM framework based on squared-loss mutual information (SMI). The proposed approach can directly handle non-linearly related objects/sequences with different dimensions, with the ability that hyper-parameters can be objectively optimized by cross-validation. We apply the proposed method to several real-world problems including image matching, unpaired voice conversion, photo album summarization, cross-feature video and cross-domain video-to-mocap alignment, and Kinect -based action recognition, and experimentally demonstrate that the proposed method is a promising alternative to state-of-the-art CDM methods.

Description: The skeleton of a 2D shape is an important geometric structure in pattern analysis and computer vision. In this paper we study the skeleton of a 2D shape in a two-manifold $mathcal {M}$ , based on a geodesic metric. We present a formal definition of the skeleton $S(Omega )$ for a shape $Omega$ in $mathcal {M}$ and show several properties that make $S(Omega )$ distinct from its Euclidean counterpart in $mathbb {R}^2$ . We further prove that for a shape sequence $lbrace Omega _irbrace$ that converge to a shape $Omega$ in $mathcal {M}$ , the mapping $Omega righta- row overline{S}(Omega )$ is lower semi-continuous. A direct application of this result is that we can use a set $P$ of sample points to approximate the boundary of a 2D shape $Omega$ in $mathcal {M}$ , and the Voronoi diagram of $P$ inside $Omega subset mathcal {M}$ gives a good approximation to the skeleton $S(Omega )$ . Examples of skeleton computation in topography and brain morphometry are illustrated.

Description: A widely used approach for locating points on deformable objects in images is to generate feature response images for each point, and then to fit a shape model to these response images. We demonstrate that Random Forest regression-voting can be used to generate high quality response images quickly. Rather than using a generative or a discriminative model to evaluate each pixel, a regressor is used to cast votes for the optimal position of each point. We show that this leads to fast and accurate shape model matching when applied in the Constrained Local Model framework. We evaluate the technique in detail, and compare it with a range of commonly used alternatives across application areas: the annotation of the joints of the hands in radiographs and the detection of feature points in facial images. We show that our approach outperforms alternative techniques, achieving what we believe to be the most accurate results yet published for hand joint annotation and state-of-the-art performance for facial feature point detection.

Description: We present a novel method to recognise planar structures in a single image and estimate their 3D orientation. This is done by exploiting the relationship between image appearance and 3D structure, using machine learning methods with supervised training data. As such, the method does not require specific features or use geometric cues, such as vanishing points. We employ general feature representations based on spatiograms of gradients and colour, coupled with relevance vector machines for classification and regression. We first show that using hand-labelled training data, we are able to classify pre-segmented regions as being planar or not, and estimate their 3D orientation. We then incorporate the method into a segmentation algorithm to detect multiple planar structures from a previously unseen image.

Description: Multiple view segmentation consists in segmenting objects simultaneously in several views. A key issue in that respect and compared to monocular settings is to ensure propagation of segmentation information between views while minimizing complexity and computational cost. In this work, we first investigate the idea that examining measurements at the projections of a sparse set of 3D points is sufficient to achieve this goal. The proposed algorithm softly assigns each of these 3D samples to the scene background if it projects on the background region in at least one view, or to the foreground if it projects on foreground region in all views. Second, we show how other modalities such as depth may be seamlessly integrated in the model and benefit the segmentation. The paper exposes a detailed set of experiments used to validate the algorithm, showing results comparable with the state of art, with reduced computational complexity. We also discuss the use of different modalities for specific situations, such as dealing with a low number of viewpoints or a scene with color ambiguities between foreground and background.

Description: Currently deep learning has made great breakthroughs in visual and speech processing, mainly because it draws lessons from the hierarchical mode that brain deals with images and speech. In the field of NLP, a topic model is one of the important ways for modeling documents. Topic models are built on a generative model that clearly does not match the way humans write. In this paper, we propose Event Model, which is unsupervised and based on the language processing mechanism of neurolinguistics, to model documents. In Event Model, documents are descriptions of concrete or abstract events seen, heard, or sensed by people and words are objects in the events. Event Model has two stages: word learning and dimensionality reduction. Word learning is to learn semantics of words based on deep learning. Dimensionality reduction is the process that representing a document as a low dimensional vector by a linear mode that is completely different from topic models. Event Model achieves state-of-the-art results on document retrieval tasks.

Description: Text detection in natural scene images is an important prerequisite for many content-based image analysis tasks, while most current research efforts only focus on horizontal or near horizontal scene text. In this paper, first we present a unified distance metric learning framework for adaptive hierarchical clustering, which can simultaneously learn similarity weights (to adaptively combine different feature similarities) and the clustering threshold (to automatically determine the number of clusters). Then, we propose an effective multi-orientation scene text detection system, which constructs text candidates by grouping characters based on this adaptive clustering. Our text candidates construction method consists of several sequential coarse-to-fine grouping steps: morphology-based grouping via single-link clustering, orientation-based grouping via divisive hierarchical clustering, and projection-based grouping also via divisive clustering. The effectiveness of our proposed system is evaluated on several public scene text databases, e.g., ICDAR Robust Reading Competition data sets (2011 and 2013), MSRA-TD500 and NEOCR. Specifically, on the multi-orientation text data set MSRA-TD500, the $f$ measure of our system is $71$ percent, much better than the state-of-the-art performance. We also construct and release a practical challenging multi-orientation scene text data set (USTB-SV1K), which is available at http://prir.ustb.edu.cn/TexStar/MOMV-text-detection/.

Description: Community detection in a complex network is an important problem of much interest in recent years. In general, a community detection algorithm chooses an objective function and captures the communities of the network by optimizing the objective function, and then, one uses various heuristics to solve the optimization problem to extract the interesting communities for the user. In this article, we demonstrate the procedure to transform a graph into points of a metric space and develop the methods of community detection with the help of a metric defined for a pair of points. We have also studied and analyzed the community structure of the network therein. The results obtained with our approach are very competitive with most of the well-known algorithms in the literature, and this is justified over the large collection of datasets. On the other hand, it can be observed that time taken by our algorithm is quite less compared to other methods and justifies the theoretical findings.

Description: A three-step iterative method with fifth-order convergence as a new modification of Newton’s method was presented. This method is for finding multiple roots of nonlinear equation with unknown multiplicity m whose multiplicity m is the highest multiplicity. Its order of convergence is analyzed and proved. Results for some numerical examples show the efficiency of the new method.

Description: To reduce the energy cost of underwater acoustic sensor networks (UWSNs), the duty cycle (i.e., periodic wake-up and sleep) concept has been used in several medium access control (MAC) protocols. Although these protocols are energy efficient, they sacrifice bandwidth utilization, which leads to lower transmission rate. In order to solve this problem, asynchronous duty cycle with network-coding Asynchronous Duty Cycle with Network-Coding MAC (ADCNC-MAC) is proposed. It contains initialization of the MAC protocol phase and data transmission phase. In the first phase, we use an asynchronous duty cycle to find a rendezvous time for exchanging data. A strategy to select network coder nodes is presented to confirm the number of network coder nodes and distribution in the network coder layer. In the data transmission phase, the network coder nodes transmit using the proposed network-coding-based algorithm and a higher volume of packet will be transmitted to the Sink with the same number of transmissions. Simulation results show that ADCNC-MAC achieves higher power efficiency, improves packet delivery ratio (PDR), and network throughput.

Description: A novel approach for event summarization and rare event detection is proposed. Unlike conventional methods that deal with event summarization and rare event detection independently, our method solves them in a single framework by transforming them into a graph editing problem. In our approach, a video is represented by a graph, each node of which indicates an event obtained by segmenting the video spatially and temporally. The edges between nodes describe the relationship between events. Based on the degree of relations, edges have different weights. After learning the graph structure, our method finds subgraphs that represent event summarization and rare events in the video by editing the graph, that is, merging its subgraphs or pruning its edges. The graph is edited to minimize a predefined energy model with the Markov Chain Monte Carlo (MCMC) method. The energy model consists of several parameters that represent the causality, frequency, and significance of events. We design a specific energy model that uses these parameters to satisfy each objective of event summarization and rare event detection. The proposed method is extended to obtain event summarization and rare event detection results across multiple videos captured from multiple views. For this purpose, the proposed method independently learns and edits each graph of individual videos for event summarization or rare event detection. Then, the method matches the extracted multiple graphs to each other, and constructs a single composite graph that represents event summarization or rare events from multiple views. Experimental results show that the proposed approach accurately summarizes multiple videos in a fully unsupervised manner . Moreover, the experiments demonstrate that the approach is advantageous in detecting rare transition of events .

Description: Object tracking has been one of the most important and active research areas in the field of computer vision. A large number of tracking algorithms have been proposed in recent years with demonstrated success. However, the set of sequences used for evaluation is often not sufficient or is sometimes biased for certain types of algorithms. Many datasets do not have common ground-truth object positions or extents, and this makes comparisons among the reported quantitative results difficult. In addition, the initial conditions or parameters of the evaluated tracking algorithms are not the same, and thus, the quantitative results reported in literature are incomparable or sometimes contradictory. To address these issues, we carry out an extensive evaluation of the state-of-the-art online object-tracking algorithms with various evaluation criteria to understand how these methods perform within the same framework. In this work, we first construct a large dataset with ground-truth object positions and extents for tracking and introduce the sequence attributes for the performance analysis. Second, we integrate most of the publicly available trackers into one code library with uniform input and output formats to facilitate large-scale performance evaluation. Third, we extensively evaluate the performance of 31 algorithms on 100 sequences with different initialization settings. By analyzing the quantitative results, we identify effective approaches for robust tracking and provide potential future research directions in this field.

Description: Fused Lasso is a popular regression technique that encodes the smoothness of the data. It has been applied successfully to many applications with a smooth feature structure. However, the computational cost of the existing solvers for fused Lasso is prohibitive when the feature dimension is extremely large. In this paper, we propose novel screening rules that are able to quickly identity the adjacent features with the same coefficients. As a result, the number of variables to be estimated can be significantly reduced, leading to substantial savings in computational cost and memory usage. To the best of our knowledge, the proposed approach is the first attempt to develop screening methods for the fused Lasso problem with general data matrix. Our major contributions are: 1) we derive a new dual formulation of fused Lasso that comes with several desirable properties; 2) we show that the new dual formulation of fused Lasso is equivalent to that of the standard Lasso by two affine transformations; 3) we propose a novel framework for developing effective and efficient screening rules for f used La sso via the m onotonicity of the s ubdifferentials (FLAMS). Some appealing features of FLAMS are: 1) our methods are safe in the sense that the detected adjacent features are guaranteed to have the same coefficients; 2) the dataset needs to be scanned only once to run the screening, whose computational cost is negligible compared to that of solving the fused Lasso; (3) FLAMS is independent of the solvers and can be integrated with any existing solvers. We have evaluated the proposed FLAMS rules on both synthetic and real datasets. The experiments indicate that FLAMS is very effective in identifying the adjacent features with the same coefficients. The speedup gained by FLAMS can be orders of magnitude.

Description: Hidden conditional random fields (HCRFs) are discriminative latent variable models which have been shown to successfully learn the hidden structure of a given classification problem. An Infinite hidden conditional random field is a hidden conditional random field with a countably infinite number of hidden states, which rids us not only of the necessity to specify a priori a fixed number of hidden states available but also of the problem of overfitting. Markov chain Monte Carlo (MCMC) sampling algorithms are often employed for inference in such models. However, convergence of such algorithms is rather difficult to verify, and as the complexity of the task at hand increases the computational cost of such algorithms often becomes prohibitive. These limitations can be overcome by variational techniques. In this paper, we present a generalized framework for infinite HCRF models, and a novel variational inference approach on a model based on coupled Dirichlet Process Mixtures, the HCRF-DPM. We show that the variational HCRF-DPM is able to converge to a correct number of represented hidden states, and performs as well as the best parametric HCRFs—chosen via cross-validation—for the difficult tasks of recognizing instances of agreement, disagreement, and pain in audiovisual sequences.

Description: In this paper, we address the challenging problem of detecting pedestrians who appear in groups. A new approach is proposed for single-pedestrian detection aided by two-pedestrian detection. A mixture model of two-pedestrian detectors is designed to capture the unique visual cues which are formed by nearby pedestrians but cannot be captured by single-pedestrian detectors. A probabilistic framework is proposed to model the relationship between the configurations estimated by single- and two-pedestrian detectors, and to refine the single-pedestrian detection result using two-pedestrian detection. The two-pedestrian detector can integrate with any single-pedestrian detector. Twenty-five state-of-the-art single-pedestrian detection approaches are combined with the two-pedestrian detector on three widely used public datasets: Caltech, TUD-Brussels, and ETH. Experimental results show that our framework improves all these approaches. The average improvement is $9$ percent on the Caltech-Test dataset, $11$ percent on the TUD-Brussels dataset and $17$ percent on the ETH dataset in terms of average miss rate. The lowest average miss rate is reduced from $37$ to percent on the Caltech-Test dataset, from $55$ to $50$ percent on the TUD-Brussels dataset and from $43$ to $38$ percent on the ETH dataset.

Description: The possibility of having information access anytime and anywhere has caused a huge increase of the popularity of wireless networks. Requirements of users and owners have been ever-increasing. However, concerns about the potential health impact of exposure to radio frequency (RF) sources have arisen and are getting accounted for in wireless network planning. In addition to adequate coverage and reduced human exposure, the installation cost of the wireless network is also an important criterion in the planning process. In this paper, a hybrid algorithm is used to optimize indoor wireless network planning while satisfying three demands: maximum coverage, minimal full installation cost (cabling, cable gutters, drilling holes, labor, etc.), and minimal human exposure. For the first time, wireless indoor networks are being optimized based on these advanced and realistic conditions. The algorithm is investigated for three scenarios and for different configurations. The impact of different exposure requirements and cost scenarios is assessed.

Description: Physical transceiver implementations for wireless communication systems usually suffer from transmit-radio frequency (Tx-RF) and receiver-RF (Rx-RF) impairments. In this paper, we aim to design efficient coordinated beamforming for multicell multiuser multi-antenna systems by fully taking into account the residual transceiver impairments. Our design objectives include both spectral efficiency and energy efficiency. In particular, we first derive the closed-form expression of the mean square error (MSE) which includes the impact of transceiver impairments. Based on that, we propose an alternating optimization algorithm to solve the coordinated multicell beamforming problems with the goal of minimizing the worst user MSE, and the sum MSE. Then, by exploiting the relationship between the minimum mean square error (MMSE) and the achievable rate, we develop a new algorithm to address the sum rate maximization problem. This approach is further generalized to solve the more intractable energy efficiency optimization problem. We prove that all the proposed iterative algorithms guarantee to converge to a stationary point. Numerical results show that our proposed schemes achieve a better performance than conventional coordinated beamforming algorithms that were designed ignoring the transceiver impairments.

Description: Systems based on bag-of-words models from image features collected at maxima of sparse interest point operators have been used successfully for both computer visual object and action recognition tasks. While the sparse, interest-point based approach to recognition is not inconsistent with visual processing in biological systems that operate in ‘saccade and fixate’ regimes, the methodology and emphasis in the human and the computer vision communities remains sharply distinct. Here, we make three contributions aiming to bridge this gap. First, we complement existing state-of-the art large scale dynamic computer vision annotated datasets like Hollywood-2  [1] and UCF Sports  [2] with human eye movements collected under the ecological constraints of visual action and scene context recognition tasks. To our knowledge these are the first large human eye tracking datasets to be collected and made publicly available for video, vision.imar.ro/eyetracking (497,107 frames, each viewed by 19 subjects), unique in terms of their (a) large scale and computer vision relevance, (b) dynamic, video stimuli, (c) task control, as well as free-viewing . Second, we introduce novel dynamic consistency and alignment measures , which underline the remarkable stability of patterns of visual search among subjects. Third, we leverage the significant amount of collected data in order to pursue studies and build automatic, end-to-end trainable computer vision systems based on human eye movements. Our studies not only shed light on the differences between computer vision spatio-temporal interest point image sampling strategies and the human fixations, as well as their impact for visual recognition performance, but also demonstrate that human fixations can be accurately predicted, and when used in an end-to-end automatic system, leveraging some of the advanced computer vision practice, can lead to state of the art results.

Description: We consider the problem of parameter estimation and energy minimization for a region-based semantic segmentation model. The model divides the pixels of an image into non-overlapping connected regions, each of which is to a semantic class. In the context of energy minimization, the main problem we face is the large number of putative pixel-to-region assignments. We address this problem by designing an accurate linear programming based approach for selecting the best set of regions from a large dictionary. The dictionary is constructed by merging and intersecting segments obtained from multiple bottom-up over-segmentations. The linear program is solved efficiently using dual decomposition. In the context of parameter estimation, the main problem we face is the lack of fully supervised data. We address this issue by developing a principled framework for parameter estimation using diverse data. More precisely, we propose a latent structural support vector machine formulation, where the latent variables model any missing information in the human annotation. Of particular interest to us are three types of annotations: (i) images segmented using generic foreground or background classes; (ii) images with bounding boxes specified for objects; and (iii) images labeled to indicate the presence of a class. Using large, publicly available datasets we show that our methods are able to significantly improve the accuracy of the region-based model.

Description: In this paper we address the problem of finding the most probable state of a discrete Markov random field (MRF), also known as the MRF energy minimization problem. The task is known to be NP-hard in general and its practical importance motivates numerous approximate algorithms. We propose a submodular relaxation approach (SMR) based on a Lagrangian relaxation of the initial problem. Unlike the dual decomposition approach of Komodakis et al. [29] SMR does not decompose the graph structure of the initial problem but constructs a submodular energy that is minimized within the Lagrangian relaxation. Our approach is applicable to both pairwise and high-order MRFs and allows to take into account global potentials of certain types. We study theoretical properties of the proposed approach and evaluate it experimentally.

Description: Higher-order Markov Random Fields, which can capture important properties of natural images, have become increasingly important in computer vision. While graph cuts work well for first-order MRF’s, until recently they have rarely been effective for higher-order MRF’s. Ishikawa’s graph cut technique [1] , [2] shows great promise for many higher-order MRF’s. His method transforms an arbitrary higher-order MRF with binary labels into a first-order one with the same minima. If all the terms are submodular the exact solution can be easily found; otherwise, pseudoboolean optimization techniques can produce an optimal labeling for a subset of the variables. We present a new transformation with better performance than [1] , [2] , both theoretically and experimentally. While [1] , [2] transforms each higher-order term independently, we use the underlying hypergraph structure of the MRF to transform a group of terms at once. For $n$ binary variables, each of which appears in terms with $k$ other variables, at worst we produce $n$ non-submodular terms, while [1] , [2] produces $O(- k)$ . We identify a local completeness property under which our method perform even better, and show that under certain assumptions several important vision problems (including common variants of fusion moves) have this property. We show experimentally that our method produces smaller weight of non-submodular edges, and that this metric is directly related to the effectiveness of QPBO [3] . Running on the same field of experts dataset used in [1] , [2] we optimally label significantly more variables (96 versus 80 percent) and converge more rapidly to a lower energy. Preliminary experiments suggest that some other higher-order MRF’s used in stereo [4] and segmentation [5] are also locally complete and would thus benefit from our work.

Description: We introduce a conceptually novel structured prediction model, GPstruct , which is kernelized, non-parametric and Bayesian, by design. We motivate the model with respect to existing approaches, among others, conditional random fields (CRFs), maximum margin Markov networks (M $^3$ N), and structured support vector machines (SVMstruct), which embody only a subset of its properties. We present an inference procedure based on Markov Chain Monte Carlo. The framework can be instantiated for a wide range of structured objects such as linear chains, trees, grids, and other general graphs. As a proof of concept, the model is benchmarked on several natural language processing tasks and a video gesture segmentation task involving a linear chain structure. We show prediction accuracies for GPstruct which are comparable to or exceeding those of CRFs and SVMstruct.

Description: In this paper, we present three improvements to a three-point third order variant of Newton’s method derived from the Simpson rule. The first one is a fifth order method using the same number of functional evaluations as the third order method, the second one is a four-point 10th order method and the last one is a five-point 20th order method. In terms of computational point of view, our methods require four evaluations (one function and three first derivatives) to get fifth order, five evaluations (two functions and three derivatives) to get 10th order and six evaluations (three functions and three derivatives) to get 20th order. Hence, these methods have efficiency indexes of 1.495, 1.585 and 1.648, respectively which are better than the efficiency index of 1.316 of the third order method. We test the methods through some numerical experiments which show that the 20th order method is very efficient.

Description: Improving the quality of degraded images is a key problem in image processing, but the breadth of the problem leads to domain-specific approaches for tasks such as super-resolution and compression artifact removal. Recent approaches have shown that a general approach is possible by learning application-specific models from examples; however, learning models sophisticated enough to generate high-quality images is computationally expensive, and so specific per-application or per-dataset models are impractical. To solve this problem, we present an efficient semi-local approximation scheme to large-scale Gaussian processes. This allows efficient learning of task-specific image enhancements from example images without reducing quality. As such, our algorithm can be easily customized to specific applications and datasets, and we show the efficiency and effectiveness of our approach across five domains: single-image super-resolution for scene, human face, and text images, and artifact removal in JPEG- and JPEG 2000-encoded images.

Description: Parsimony, including sparsity and low rank, has been shown to successfully model data in numerous machine learning and signal processing tasks. Traditionally, such modeling approaches rely on an iterative algorithm that minimizes an objective function with parsimony-promoting terms. The inherently sequential structure and data-dependent complexity and latency of iterative optimization constitute a major limitation in many applications requiring real-time performance or involving large-scale data. Another limitation encountered by these modeling techniques is the difficulty of their inclusion in discriminative learning scenarios. In this work, we propose to move the emphasis from the model to the pursuit algorithm, and develop a process-centric view of parsimonious modeling, in which a learned deterministic fixed-complexity pursuit process is used in lieu of iterative optimization. We show a principled way to construct learnable pursuit process architectures for structured sparse and robust low rank models, derived from the iteration of proximal descent algorithms. These architectures learn to approximate the exact parsimonious representation at a fraction of the complexity of the standard optimization methods. We also show that appropriate training regimes allow to naturally extend parsimonious models to discriminative settings. State-of-the-art results are demonstrated on several challenging problems in image and audio processing with several orders of magnitude speed-up compared to the exact optimization algorithms.

Description: CANDECOMP/PARAFAC (CP) tensor factorization of incomplete data is a powerful technique for tensor completion through explicitly capturing the multilinear latent factors. The existing CP algorithms require the tensor rank to be manually specified, however, the determination of tensor rank remains a challenging problem especially for CP rank . In addition, existing approaches do not take into account uncertainty information of latent factors, as well as missing entries. To address these issues, we formulate CP factorization using a hierarchical probabilistic model and employ a fully Bayesian treatment by incorporating a sparsity-inducing prior over multiple latent factors and the appropriate hyperpriors over all hyperparameters, resulting in automatic rank determination. To learn the model, we develop an efficient deterministic Bayesian inference algorithm, which scales linearly with data size. Our method is characterized as a tuning parameter-free approach, which can effectively infer underlying multilinear factors with a low-rank constraint, while also providing predictive distributions over missing entries. Extensive simulations on synthetic data illustrate the intrinsic capability of our method to recover the ground-truth of CP rank and prevent the overfitting problem, even when a large amount of entries are missing. Moreover, the results from real-world applications, including image inpainting and facial image synthesis, demonstrate that our method outperforms state-of-the-art approaches for both tensor factorization and tensor completion in terms of predictive performance.

Description: Existing deep convolutional neural networks (CNNs) require a fixed-size (e.g., 224 $times$ 224) input image. This requirement is “artificial” and may reduce the recognition accuracy for the images or sub-images of an arbitrary size/scale. In this work, we equip the networks with another pooling strategy, “spatial pyramid pooling”, to eliminate the above requirement. The new network structure, called SPP-net, can generate a fixed-length representation regardless of image size/scale. Pyramid pooling is also robust to object deformations. With these advantages, SPP-net should in general improve all CNN-based image classification methods. On the ImageNet 2012 dataset, we demonstrate that SPP-net boosts the accuracy of a variety of CNN architectures despite their different designs. On the Pascal VOC 2007 and Caltech101 datasets, SPP-net achieves state-of-the-art classification results using a single full-image representation and no fine-tuning. The power of SPP-net is also significant in object detection. Using SPP-net, we compute the feature maps from the entire image only once, and then pool features in arbitrary regions (sub-images) to generate fixed-length representations for training the detectors. This method avoids repeatedly computing the convolutional features. In processing test images, our method is 24-102 $times$ faster than the R-CNN method, while achieving better or comparable accuracy on Pascal VOC 2007. In ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2014, our methods rank #2 in object detection and #3 in image classification among all 38 teams. This manuscript also introduces the improvement made for this - ompetition.

Description: We present efficient graph cut algorithms for three problems: (1) finding a region in an image, so that the histogram (or distribution) of an image feature within the region most closely matches a given model; (2) co-segmentation of image pairs and (3) interactive image segmentation with a user-provided bounding box. Each algorithm seeks the optimum of a global cost function based on the Bhattacharyya measure, a convenient alternative to other matching measures such as the Kullback–Leibler divergence. Our functionals are not directly amenable to graph cut optimization as they contain non-linear functions of fractional terms, which make the ensuing optimization problems challenging. We first derive a family of parametric bounds of the Bhattacharyya measure by introducing an auxiliary labeling. Then, we show that these bounds are auxiliary functions of the Bhattacharyya measure, a result which allows us to solve each problem efficiently via graph cuts. We show that the proposed optimization procedures converge within very few graph cut iterations. Comprehensive and various experiments, including quantitative and comparative evaluations over two databases, demonstrate the advantages of the proposed algorithms over related works in regard to optimality, computational load, accuracy and flexibility.

Description: Robust small target detection of low signal-to-noise ratio (SNR) is very important in infrared search and track applications for self-defense or attacks. Due to the complex background, current algorithms have some unsolved issues with false alarm rate. In order to reduce the false alarm rate, an infrared small target detection algorithm based on saliency detection and support vector machine was proposed. Firstly, we detect salient regions that may contain targets with phase spectrum Fourier transform (PFT) approach. Then, target recognition was performed in the salient regions. Experimental results show the proposed algorithm has ideal robustness and efficiency for real infrared small target detection applications.

Description: In dynamic propagation environments, beamforming algorithms may suffer from strong interference, steering vector mismatches, a low convergence speed and a high computational complexity. Reduced-rank signal processing techniques provide a way to address the problems mentioned above. This paper presents a low-complexity robust data-dependent dimensionality reduction based on an iterative optimization with steering vector perturbation (IOVP) algorithm for reduced-rank beamforming and steering vector estimation. The proposed robust optimization procedure jointly adjusts the parameters of a rank reduction matrix and an adaptive beamformer. The optimized rank reduction matrix projects the received signal vector onto a subspace with lower dimension. The beamformer/steering vector optimization is then performed in a reduced dimension subspace. We devise efficient stochastic gradient and recursive least-squares algorithms for implementing the proposed robust IOVP design. The proposed robust IOVP beamforming algorithms result in a faster convergence speed and an improved performance. Simulation results show that the proposed IOVP algorithms outperform some existing full-rank and reduced-rank algorithms with a comparable complexity.

Description: Recently, wireless sensor networks (WSNs) have drawn great interest due to their outstanding monitoring and management potential in medical, environmental and industrial applications. Most of the applications that employ WSNs demand all of the sensor nodes to run on a common time scale, a requirement that highlights the importance of clock synchronization. The clock synchronization problem in WSNs is inherently related to parameter estimation. The accuracy of clock synchronization algorithms depends essentially on the statistical properties of the parameter estimation algorithms. Recently, studies dedicated to the estimation of synchronization parameters, such as clock offset and skew, have begun to emerge in the literature. The aim of this article is to provide an overview of the state-of-the-art clock synchronization algorithms for WSNs from a statistical signal processing point of view. This article focuses on describing the key features of the class of clock synchronization algorithms that exploit the traditional two-way message (signal) exchange mechanism. Upon introducing the two-way message exchange mechanism, the main clock offset estimation algorithms for pairwise synchronization of sensor nodes are first reviewed, and their performance is compared. The class of fully-distributed clock offset estimation algorithms for network-wide synchronization is then surveyed. The paper concludes with a list of open research problems pertaining to clock synchronization of WSNs.

Description: Network coding is an emerging technique known to improve the network performance in many aspects. In Vehicular Ad-hoc Networks (VANET), the bandwidth is considered to be one of the most important network resources. In this paper, we propose a network coding technique to improve the bandwidth utilization for non-safety applications in VANET. In a scenario where there are two sources broadcasting the data into the same area at the same time, the relay will use the network coding technique to decrease the number of rebroadcasting events and the consumption of the bandwidth, However, a fundamental problem for the relay when it receives a packet, is whether to wait for a coding opportunity and save the bandwidth or send the packet directly and reduce the delay. In order to address such tradeoff, we introduce two versions of our protocol, namely buffer size control scheme (BSCS) and time control scheme (TCS); by both versions we aim to control the delay that is experienced by the packet at each hop, while achieving better bandwidth utilization.Up to 38 % improvement in the bandwidth utilization has been recorded, and both schemes have shown a considerable amount of control on the imposed delay.

Description: Long-Term Evolution (LTE) was implemented to fulfill and satisfy users’ needs as well as their demands for an improvised, fast and efficient Quality of service (QoS). A minimal aggregate of waiting time in return would give users a better Quality of experience (QoE). Real-time service packet scheduling is an important process in allocating resources to users. An efficient packet scheduling scheme will be able to cater fairly and efficiently to its users in the LTE network. Hence, studies are performed focusing on real-time traffic which includes video as well as Voice over Internet Protocol (VoIP) transmissions. In this work, the existing exponential rule (EXP rule) is utilized to benchmark our proposed packet scheduling techniques so that we are able to further evaluate the scheduling performance. In response to the increasing likelihood of losing packets in the EXP rule’s algorithm and maximizing the throughput rate, several schemes have been experimented with. The proposed schemes include 1) simplified EXP rule (sEXP Rule), 2) modified EXP rule (mEXP Rule), 3) EXP rule with maximum throughput (MT) (EXP_MT Rule), and 4) enhanced EXP rule with MT (E2M). By adding MT as a weight to the EXP rule, the throughput is maximized, thus providing higher throughput rates for real-time and non-real-time traffic. The simulation results show that the sEXP rule has a better performance in throughput, packet loss rate (PLR), and spectral efficiency for video traffic. Aside from this, our proposed E2M rule performs better than the benchmark EXP rule and outperforms the other proposed schemes, as well. It is observed that the E2M rule has better QoS support for real-time transmission in terms of delay, packet loss, throughput and spectral efficiency, within the LTE network. Hence, our proposed E2M rule is an enhancement of the benchmark EXP rule, which is commonly used in LTE packet scheduling.

Description: In delay tolerant networks (DTNs), the network may not be fully connected at any instant of time, but connections occurring between nodes at different times make the network connected through the entire time continuum. In such a case, traditional routing methods fail to operate because there are no contemporaneous end-to-end paths between sources and destinations. This study examines the routing in DTNs where connections arise in a periodic nature. We analyze various levels of periodicity in order to meet the requirements of different network models. We propose different routing algorithms for different kinds of periodic connections. Our proposed routing methods guarantee the earliest delivery time and minimum hop-count, simultaneously. We evaluate our routing schemes via extensive simulation experiments and compare them to some other popular routing approaches proposed for DTNs. Our evaluations show the feasibility and effectiveness of our schemes as viable routing methods for delay tolerant networks.

Description: Due to rapid developments in mobile technology as well as various multimedia features like messaging, browsing, and streaming, user-created mobile contents are increasing, both in terms of quantity and quality, and at the same time are shared in real time. To get into step with such movements, new content-centric networking (CCN) has appeared. However, CCN has not taken the effect of consumer device movements into consideration. So, this paper proposes a partial path extension scheme to provide lower communication overhead, shorter download time, and lower network resource consumption in mobile consumer environments.

Description: In this paper we investigate some parallel variants of Broyden’s method and, for the basic variant, we present its convergence properties. The main result is that the behavior of the considered parallel Broyden’s variants is comparable with the classical parallel Newton method, and significantly better than the parallel Cimmino method, both for linear and nonlinear cases. The considered variants are also compared with two more recently proposed parallel Broyden’s method. Some numerical experiments are presented to illustrate the advantages and limits of the proposed algorithms.

Description: The sign least mean square with reweighted L1-norm constraint (SLMS-RL1) algorithm is an attractive sparse channel estimation method among Gaussian mixture model (GMM) based algorithms for use in impulsive noise environments. The channel sparsity can be exploited by SLMS-RL1 algorithm based on appropriate reweighted factor, which is one of key parameters to adjust the sparse constraint for SLMS-RL1 algorithm. However, to the best of the authors’ knowledge, a reweighted factor selection scheme has not been developed. This paper proposes a Monte-Carlo (MC) based reweighted factor selection method to further strengthen the performance of SLMS-RL1 algorithm. To validate the performance of SLMS-RL1 using the proposed reweighted factor, simulations results are provided to demonstrate that convergence speed can be reduced by increasing the channel sparsity, while the steady-state MSE performance only slightly changes with different GMM impulsive-noise strengths.

Description: In modern mobile telecommunications, shadow fading has to be modeled by a two-dimensional (2D) correlated random variable since shadow fading may present both cross-correlation and spatial correlation due to the presence of similar obstacles during the propagation. In this paper, 2D correlated random shadowing is generated based on the multi-resolution frequency domain ParFlow (MR-FDPF) model. The MR-FDPF model is a 2D deterministic radio propagation model, so a 2D deterministic shadowing can be firstly extracted from it. Then, a 2D correlated random shadowing can be generated by considering the extracted 2D deterministic shadowing to be a realization of it. Moreover, based on the generated 2D correlated random shadowing, a complete 2D semi-deterministic path loss model can be proposed. The proposed methodology of this paper can be implemented into system-level simulators where it will be very useful due to its ability to generate realistic shadow fading.

Description: We present a local convergence analysis of an eighth order three step methodin order to approximate a locally unique solution of nonlinear equation in a Banach spacesetting. In an earlier study by Sharma and Arora (2015), the order of convergence wasshown using Taylor series expansions and hypotheses up to the fourth order derivative oreven higher of the function involved which restrict the applicability of the proposed scheme.However, only first order derivative appears in the proposed scheme. In order to overcomethis problem, we proposed the hypotheses up to only the first order derivative. In this way,we not only expand the applicability of the methods but also propose convergence domain.Finally, where earlier studies cannot be applied, a variety of concrete numerical examplesare proposed to obtain the solutions of nonlinear equations. Our study does not exhibit thistype of problem/restriction.

Description: Lung cancer continues to rank as the leading cause of cancer deaths worldwide. One of the most promising techniques for early detection of cancerous cells relies on sputum cell analysis. This was the motivation behind the design and the development of a new computer aided diagnosis (CAD) system for early detection of lung cancer based on the analysis of sputum color images. The proposed CAD system encompasses four main processing steps. First is the preprocessing step which utilizes a Bayesian classification method using histogram analysis. Then, in the second step, mean shift segmentation is applied to segment the nuclei from the cytoplasm. The third step is the feature analysis. In this step, geometric and chromatic features are extracted from the nucleus region. These features are used in the diagnostic process of the sputum images. Finally, the diagnosis is completed using an artificial neural network and support vector machine (SVM) for classifying the cells into benign or malignant. The performance of the system was analyzed based on different criteria such as sensitivity, specificity and accuracy. The evaluation was carried out using Receiver Operating Characteristic (ROC) curve. The experimental results demonstrate the efficiency of the SVM classifier over other classifiers, with 97% sensitivity and accuracy as well as a significant reduction in the number of false positive and false negative rates.

Description: Near-infrared spectroscopy (NIRS) enables the non-invasive measurement of changes in hemodynamics and oxygenation in tissue. Changes in light-coupling due to movement of the subject can cause movement artifacts (MAs) in the recorded signals. Several methods have been developed so far that facilitate the detection and reduction of MAs in the data. However, due to fixed parameter values (e.g., global threshold) none of these methods are perfectly suitable for long-term (i.e., hours) recordings or were not time-effective when applied to large datasets. We aimed to overcome these limitations by automation, i.e., data adaptive thresholding specifically designed for long-term measurements, and by introducing a stable long-term signal reconstruction. Our new technique (“acceleration-based movement artifact reduction algorithm”, AMARA) is based on combining two methods: the “movement artifact reduction algorithm” (MARA, Scholkmann et al. Phys. Meas. 2010, 31, 649–662), and the “accelerometer-based motion artifact removal” (ABAMAR, Virtanen et al. J. Biomed. Opt. 2011, 16, 087005). We describe AMARA in detail and report about successful validation of the algorithm using empirical NIRS data, measured over the prefrontal cortex in adolescents during sleep. In addition, we compared the performance of AMARA to that of MARA and ABAMAR based on validation data.

Description: This paper focuses on the parameter identification problem for Wiener nonlinear dynamic systems with moving average noises. In order to improve the convergence rate, the gradient-based iterative algorithm is presented by replacing the unmeasurable variables with their corresponding iterative estimates, and to compute iteratively the noise estimates based on the obtained parameter estimates. The simulation results show that the proposed algorithm can effectively estimate the parameters of Wiener systems with moving average noises.

Description: In this paper, the dynamical behavior of different optimal iterative schemes for solving nonlinear equations with increasing order, is studied. The tendency of the complexity of the Julia set is analyzed and referred to the fractal dimension. In fact, this fractal dimension can be shown to be a powerful tool to compare iterative schemes that estimate the solution of a nonlinear equation. Based on the box-counting algorithm, several iterative derivative-free methods of different convergence orders are compared.

Description: The basis of vehicular ad hoc networks (VANETs) is the exchange of data between entities, and making a decision on received data/event is usually based on information provided by other entities. Many researchers utilize the concept of trust to assess the trustworthiness of the received data. Nevertheless, the lack of a review to sum up the best available research on specific questions on trust management in vehicular ad hoc networks is sensible. This paper presents a systematic literature review to provide comprehensive and unbiased information about various current trust conceptions, proposals, problems, and solutions in VANETs to increase quality of data in transportation. For the purpose of the writing of this paper, a total of 111 articles related to the trust model in VANETs published between 2005 and 2014 were extracted from the most relevant scientific sources (IEEE Computer Society, ACM Digital Library, Springer Link, Science Direct, and Wiley Online Library). Finally, ten articles were eventually analyzed due to several reasons such as relevancy and comprehensiveness of discussion presented in the articles. Using the systematic method of review, this paper succeeds to reveal the main challenges and requirements for trust in VANETs and future research within this scope.

Description: Hardware imperfections can significantly reduce the performance of full-duplex wireless systems by introducing non-idealities and random effects that make it challenging to fully suppress self-interference. Previous research has mostly focused on analysing the impact of hardware imperfections on full-duplex systems, based on simulations and theoretical models. In this paper, we follow a measurement-based approach to experimentally identify and isolate these hardware imperfections leading to residual self-interference in full-duplex nodes. Our measurements show the important role of images arising from in-phase and quadrature (IQ) imbalance in the transmitter and receiver mixers. We also observe baseband non-linearities in the digital-to-analog converters (DAC), which can introduce strong harmonic components in the transmitted signal that have not been considered previously. A corresponding general mathematical model to suppress these components of the self-interference signal arising from the hardware non-idealities is developed from the observations and measurements. Results from a 10 MHz bandwidth full-duplex system, operating at 2.48 GHz, show that up to 13 dB additional suppression, relative to state-of-the-art implementations, can be achieved by jointly compensating for IQ imbalance and DAC non-linearities.

Description: Recent years have witnessed the increasing efforts toward making architecture standardization for the secured wireless mobile ad hoc networks. In this scenario when a node actively utilizes the other node resources for communicating and refuses to help other nodes in their transmission or reception of data, it is called a selfish node. As the entire mobile ad hoc network (MANETs) depends on cooperation from neighboring nodes, it is very important to detect and eliminate selfish nodes from being part of the network. In this paper, token-based umpiring technique (TBUT) is proposed, where every node needs a token to participate in the network and the neighboring nodes act as umpire. This proposed TBUT is found to be very efficient with a reduced detection time and less overhead. The security analysis and experimental results have shown that TBUT is feasible for enhancing the security and network performance of real applications.

Description: In this paper, multiple device-to-device (D2D) communication underlaying cellular multiuser multiple inputs multiple outputs (MU-MIMO) systems is investigated. This type of communication can improve spectral efficiency to address future demand, but interference management, user clustering, and resource allocation are three key problems related to resource sharing. Interference alignment (IA) is proposed to better mitigate in-cluster interference compared with a multiplex scheme, and user clustering and resource allocation are jointly investigated using binary-integer programming. In addition to an exhaustive search for a maximum throughput, we propose a two-step suboptimal algorithm by reducing the search space and applying branch-and-bound searching (BBS). To further obtain a good trade-off between performance and complexity, we propose a novel algorithm based on distance-constrained criteria for user clustering. The simulation results show that the IA and multiplex schemes acquiring user clustering gains outperform the orthogonal scheme without user clustering. Besides, the proposed two-step and location-based algorithms achieve little losses compared with the optimal algorithm under low complexities.

Description: When a wireless sensor network (WSN) is employed to monitor environmental data, efficient energy usage is demanded so that the WSN lifetime can be extended. In this context, we propose a distributed energy con...

Description: Recent years, information spreading under the environment of wireless communication has attracted increasing interest. Microblog platform on mobile terminals, as one product of wireless communication, facilita...

Description: Internet of Thing (IoT) or also referred to as IP-enabled wireless sensor network (IP-WSN) has become a rich area of research. This is due to the rapid growth in a wide spectrum of critical application domains...

Description: Throughput imbalances among contending flows are known to occur when any carrier sense multiple access (CSMA)-based protocol is employed in multi-hop wireless networks. These imbalances may vary from slight di...

Description: Clustering is a fundamental task in data mining. Affinity propagation clustering (APC) is an effective and efficient clustering technique that has been applied in various domains. APC iteratively propagates information between affinity samples, updates the responsibility matrix and availability matrix, and employs these matrices to choose cluster centers (or exemplars) of respective clusters. However, since it mainly uses negative Euclidean distance between exemplars and samples as the similarity between them, it is difficult to identify clusters with complex structure. Therefore, the performance of APC deteriorates on samples distributed with complex structure. To mitigate this problem, we propose an improved APC based on a path-based similarity (APC-PS). APC-PS firstly utilizes negative Euclidean distance to find exemplars of clusters. Then, it employs the path-based similarity to measure the similarity between exemplars and samples, and to explore the underlying structure of clusters. Next, it assigns non-exemplar samples to their respective clusters via that similarity. Our empirical study on synthetic and UCI datasets shows that the proposed APC-PS significantly outperforms original APC and other related approaches.

Description: Graph-based semi-supervised classification uses a graph to capture the relationship between samples and exploits label propagation techniques on the graph to predict the labels of unlabeled samples. However, it is difficult to construct a graph that faithfully describes the relationship between high-dimensional samples. Recently, low-rank representation has been introduced to construct a graph, which can preserve the global structure of high-dimensional samples and help to train accurate transductive classifiers. In this paper, we take advantage of low-rank representation for graph construction and propose an inductive semi-supervised classifier called Semi-Supervised Classification based on Low-Rank Representation (SSC-LRR). SSC-LRR first utilizes a linearized alternating direction method with adaptive penalty to compute the coefficient matrix of low-rank representation of samples. Then, the coefficient matrix is adopted to define a graph. Finally, SSC-LRR incorporates this graph into a graph-based semi-supervised linear classifier to classify unlabeled samples. Experiments are conducted on four widely used facial datasets to validate the effectiveness of the proposed SSC-LRR and the results demonstrate that SSC-LRR achieves higher accuracy than other related methods.

Description: This research proposes a two-stage user-based collaborative filtering process using an artificial immune system for the prediction of student grades, along with a filter for professor ratings in the course recommendation for college students. We test for cosine similarity and Karl Pearson (KP) correlation in affinity calculations for clustering and prediction. This research uses student information and professor information datasets of Yuan Ze University from the years 2005–2009 for the purpose of testing and training. The mean average error and confusion matrix analysis form the testing parameters. A minimum professor rating was tested to check the results, and observed that the recommendation systems herein provide highly accurate results for students with higher mean grades.

Description: In this paper, we investigate the coexistence of two technologies that have been put forward for the fifth generation (5G) of cellular networks, namely, network-assisted device-to-device (D2D) communications a...

Description: Cognitive radio networks (CRN), in their quest to become the preferred next-generation wireless communication paradigm, will depend heavily on their ability to efficiently manage the limited resources at their...

Description: This paper is concerned with the application of computational intelligence techniques to the conceptual design and development of a large-scale floating settlement. The settlement in question is a design for the area of Urla, which is a rural touristic region located on the west coast of Turkey, near the metropolis of Izmir. The problem at hand includes both engineering and architectural aspects that need to be addressed in a comprehensive manner. We thus adapt the view as a multi-objective constrained real-parameter optimization problem. Specifically, we consider three objectives, which are conflicting. The first one aims at maximizing accessibility of urban functions such as housing and public spaces, as well as special functions, such as a marina for yachts and a yacht club. The second one aims at ensuring the wind protection of the general areas of the settlement, by adequately placing them in between neighboring land masses. The third one aims at maximizing visibility of the settlement from external observation points, so as to maximize the exposure of the settlement. To address this complex multi-objective optimization problem and identify lucrative alternative design solutions, a multi-objective harmony search algorithm (MOHS) is developed and applied in this paper. When compared to the Differential Evolution algorithm developed for the problem in the literature, we demonstrate that MOHS achieves competitive or slightly better performance in terms of hyper volume calculation, and gives promising results when the Pareto front approximation is examined.

Description: By modulating the electric field induced to a human body, it is possible to transfer data wirelessly using the body as a transmission medium. This method is referred to as human body communication (HBC), which...

Description: Provides a listing of board members, committee members, editors, and society officers.

Description: Certain inner feelings and physiological states like pain are subjective states that cannot be directly measured, but can be estimated from spontaneous facial expressions. Since they are typically characterized by subtle movements of facial parts, analysis of the facial details is required. To this end, we formulate a new regression method for continuous estimation of the intensity of facial behavior interpretation, called Doubly Sparse Relevance Vector Machine (DSRVM). DSRVM enforces double sparsity by jointly selecting the most relevant training examples (a.k.a. relevance vectors) and the most important kernels associated with facial parts relevant for interpretation of observed facial expressions. This advances prior work on multi-kernel learning, where sparsity of relevant kernels is typically ignored. Empirical evaluation on challenging Shoulder Pain videos, and the benchmark DISFA and SEMAINE datasets demonstrate that DSRVM outperforms competing approaches with a multi-fold reduction of running times in training and testing.

Description: In network virtualization, one of its core challenges lies in how to map the virtual networks (VNs) to the shared substrate network (SN) that is managed by an infrastructure provider, termed as the virtual net...

Description: Recent years have witnessed major innovations in mobile crowdsourcing networks. For selfish participants, conventional methods resort to incentive mechanism design for resource utilization, which might overloo...

Description: During a construction project life cycle, project costs and time estimations contribute greatly to baseline scheduling. Besides, schedule risk analysis and project control are also influenced by the above factors. Although many papers have offered estimation techniques, little attempt has been made to generate project time series data as daily progressive estimations in different project environments that could help researchers in generating general and customized formulae in further studies. This paper, however, is an attempt to introduce a new simulation approach to reflect the data regarding time series progress of the project, considering the specifications and the complexity of the project and the environment where the project is performed. Moreover, this simulator can equip project managers with estimated information, which reassures them of the execution stages of the project although they lack historical data. A case study is presented to show the usefulness of the model and its applicability in practice. In this study, singular spectrum analysis has been employed to analyze the simulated outputs, and the results are separated based on their signal and noise trends. The signal trend is used as a point-of-reference to compare the outputs of a simulation employing S-curve technique results and the formulae corresponding to earned value management, as well as the life of a given project.

Description: This paper discusses the parameter estimation problems of multi-input output-error autoregressive (OEAR) systems. By combining the auxiliary model identification idea and the data filtering technique, a data filtering based recursive generalized least squares (F-RGLS) identification algorithm and a data filtering based iterative least squares (F-LSI) identification algorithm are derived. Compared with the F-RGLS algorithm, the proposed F-LSI algorithm is more effective and can generate more accurate parameter estimates. The simulation results confirm this conclusion.

Description: We propose a one-shot (non-iterative) cooperative beamforming scheme for downlink multicell systems. Unlike previous non-iterative beamforming schemes, the proposed cooperative beamforming strives to balance m...

Description: Intra-vehicular wireless sensor network (IVWSN) enables the integration of the wireless sensor network technology into the vehicle architecture through either eliminating the wires between the existing sensors...

Description: The force-directed paradigm is one of the few generic approaches to drawing graphs. Since force-directed algorithms can be extended easily, they are used frequently. Most of these algorithms are, however, quite slow on large graphs, as they compute a quadratic number of forces in each iteration. We give a new algorithm that takes only O ( m + n log n ) time per iteration when laying out a graph with n vertices and m edges. Our algorithm approximates the true forces using the so-called well-separated pair decomposition. We perform experiments on a large number of graphs and show that we can strongly reduce the runtime, even on graphs with less than a hundred vertices, without a significant influence on the quality of the drawings (in terms of the number of crossings and deviation in edge lengths).

Description: Raindrops adhered to a windscreen or window glass can significantly degrade the visibility of a scene. Modeling, detecting and removing raindrops will, therefore, benefit many computer vision applications, particularly outdoor surveillance systems and intelligent vehicle systems. In this paper, a method that automatically detects and removes adherent raindrops is introduced. The core idea is to exploit the local spatio-temporal derivatives of raindrops. To accomplish the idea, we first model adherent raindrops using law of physics, and detect raindrops based on these models in combination with motion and intensity temporal derivatives of the input video. Having detected the raindrops, we remove them and restore the images based on an analysis that some areas of raindrops completely occludes the scene, and some other areas occlude only partially. For partially occluding areas, we restore them by retrieving as much as possible information of the scene, namely, by solving a blending function on the detected partially occluding areas using the temporal intensity derivative. For completely occluding areas, we recover them by using a video completion technique. Experimental results using various real videos show the effectiveness of our method.

Description: Deep convolutional networks have proven to be very successful in learning task specific features that allow for unprecedented performance on various computer vision tasks. Training of such networks follows mostly the supervised learning paradigm, where sufficiently many input-output pairs are required for training. Acquisition of large training sets is one of the key challenges, when approaching a new task. In this paper, we aim for generic feature learning and present an approach for training a convolutional network using only unlabeled data. To this end, we train the network to discriminate between a set of surrogate classes. Each surrogate class is formed by applying a variety of transformations to a randomly sampled ‘seed’ image patch. In contrast to supervised network training, the resulting feature representation is not class specific. It rather provides robustness to the transformations that have been applied during training. This generic feature representation allows for classification results that outperform the state of the art for unsupervised learning on several popular datasets (STL-10, CIFAR-10, Caltech-101, Caltech-256). While features learned with our approach cannot compete with class specific features from supervised training on a classification task, we show that they are advantageous on geometric matching problems, where they also outperform the SIFT descriptor.

Description: We address the video-based face association problem, in which one attempts to extract the face tracks of multiple subjects while maintaining label consistency. Traditional tracking algorithms have difficulty in handling this task, especially when challenging nuisance factors like motion blur, low resolution or significant camera motions are present. We demonstrate that contextual features, in addition to face appearance itself, play an important role in this case. We propose principled methods to combine multiple features using Conditional Random Fields and Max-Margin Markov networks to infer labels for the detected faces. Different from many existing approaches, our algorithms work in online mode and hence have a wider range of applications. We address issues such as parameter learning, inference and handling false positves/negatives that arise in the proposed approach. Finally, we evaluate our approach on several public databases.

Description: There are many scenarios in artificial intelligence, signal processing or medicine, in which a temporal sequence consists of several unknown overlapping independent causes, and we are interested in accurately recovering those canonical causes. Factorial hidden Markov models (FHMMs) present the versatility to provide a good fit to these scenarios. However, in some scenarios, the number of causes or the number of states of the FHMM cannot be known or limited a priori. In this paper, we propose an infinite factorial unbounded-state hidden Markov model (IFUHMM), in which the number of parallel hidden Markovmodels (HMMs) and states in each HMM are potentially unbounded. We rely on a Bayesian nonparametric (BNP) prior over integer-valued matrices, in which the columns represent the Markov chains, the rows the time indexes, and the integers the state for each chain and time instant. First, we extend the existent infinite factorial binary-state HMM to allow for any number of states. Then, we modify this model to allow for an unbounded number of states and derive an MCMC-based inference algorithm that properly deals with the trade-off between the unbounded number of states and chains. We illustrate the performance of our proposed models in the power disaggregation problem.

Description: The speed with which intelligent systems can react to an action depends on how soon it can be recognized. The ability to recognize ongoing actions is critical in many applications, for example, spotting criminal activity. It is challenging, since decisions have to be made based on partial videos of temporally incomplete action executions. In this paper, we propose a novel discriminative multi-scale kernelized model for predicting the action class from a partially observed video. The proposed model captures temporal dynamics of human actions by explicitly considering all the history of observed features as well as features in smaller temporal segments. A compositional kernel is proposed to hierarchically capture the relationships between partial observations as well as the temporal segments, respectively. We develop a new learning formulation, which elegantly captures the temporal evolution over time, and enforces the label consistency between segments and corresponding partial videos. We prove that the proposed learning formulation minimizes the upper bound of the empirical risk. Experimental results on four public datasets show that the proposed approach outperforms state-of-the-art action prediction methods.

Description: Evidence is mounting that Convolutional Networks (ConvNets) are the most effective representation learning method for visual recognition tasks. In the common scenario, a ConvNet is trained on a large labeled dataset (source) and the feed-forward units activation of the trained network, at a certain layer of the network, is used as a generic representation of an input image for a task with relatively smaller training set (target). Recent studies have shown this form of representation transfer to be suitable for a wide range of target visual recognition tasks. This paper introduces and investigates several factors affecting the transferability of such representations. It includes parameters for training of the source ConvNet such as its architecture, distribution of the training data, etc. and also the parameters of feature extraction such as layer of the trained ConvNet, dimensionality reduction, etc. Then, by optimizing these factors, we show that significant improvements can be achieved on various (17) visual recognition tasks. We further show that these visual recognition tasks can be categorically ordered based on their similarity to the source task such that a correlation between the performance of tasks and their similarity to the source task w.r.t. the proposed factors is observed.

Description: Graph matching (GM) is a fundamental problem in computer science, and it plays a central role to solve correspondence problems in computer vision. GM problems that incorporate pairwise constraints can be formulated as a quadratic assignment problem (QAP). Although widely used, solving the correspondence problem through GM has two main limitations: (1) the QAP is NP-hard and difficult to approximate; (2) GM algorithms do not incorporate geometric constraints between nodes that are natural in computer vision problems. To address aforementioned problems, this paper proposes factorized graph matching (FGM). FGM factorizes the large pairwise affinity matrix into smaller matrices that encode the local structure of each graph and the pairwise affinity between edges. Four are the benefits that follow from this factorization: (1) There is no need to compute the costly (in space and time) pairwise affinity matrix; (2) The factorization allows the use of a path-following optimization algorithm, that leads to improved optimization strategies and matching performance; (3) Given the factorization, it becomes straight-forward to incorporate geometric transformations (rigid and non-rigid) to the GM problem. (4) Using a matrix formulation for the GM problem and the factorization, it is easy to reveal commonalities and differences between different GM methods. The factorization also provides a clean connection with other matching algorithms such as iterative closest point; Experimental results on synthetic and real databases illustrate how FGM outperforms state-of-the-art algorithms for GM. The code is available at http://humansensing.cs.cmu.edu/fgm .

Description: Existing data association techniques mostly focus on matching pairs of data-point sets and then repeating this process along space-time to achieve long term correspondences. However, in many problems such as person re-identification, a set of data-points may be observed at multiple spatio-temporal locations and/or by multiple agents in a network and simply combining the local pairwise association results between sets of data-points often leads to inconsistencies over the global space-time horizons. In this paper, we propose a Novel Network Consistent Data Association (NCDA) framework formulated as an optimization problem that not only maintains consistency in association results across the network, but also improves the pairwise data association accuracies. The proposed NCDA can be solved as a binary integer program leading to a globally optimal solution and is capable of handling the challenging data-association scenario where the number of data-points varies across different sets of instances in the network. We also present an online implementation of NCDA method that can dynamically associate new observations to already observed data-points in an iterative fashion, while maintaining network consistency. We have tested both the batch and the online NCDA in two application areas—person re-identification and spatio-temporal cell tracking and observed consistent and highly accurate data association results in all the cases.

Description: This paper presents a method for learning an And-Or model to represent context and occlusion for car detection and viewpoint estimation. The learned And-Or model represents car-to-car context and occlusion configurations at three levels: (i) spatially-aligned cars, (ii) single car under different occlusion configurations, and (iii) a small number of parts. The And-Or model embeds a grammar for representing large structural and appearance variations in a reconfigurable hierarchy. The learning process consists of two stages in a weakly supervised way (i.e., only bounding boxes of single cars are annotated). First, the structure of the And-Or model is learned with three components: (a) mining multi-car contextual patterns based on layouts of annotated single car bounding boxes, (b) mining occlusion configurations between single cars, and (c) learning different combinations of part visibility based on CAD simulations. The And-Or model is organized in a directed and acyclic graph which can be inferred by Dynamic Programming. Second, the model parameters (for appearance, deformation and bias) are jointly trained using Weak-Label Structural SVM. In experiments, we test our model on four car detection datasets—the KITTI dataset [1] , the PASCAL VOC2007 car dataset  [2] , and two self-collected car datasets, namely the Street-Parking car dataset and the Parking-Lot car dataset, and three datasets for car viewpoint estimation—the PASCAL VOC2006 car dataset  [2] , the 3D car dataset  [3] , and the PASCAL3D+ car dataset  [4] . Compared with state-of-the-art variants of deformable part-based models and other methods, our model achieves significant improvement consistently on the four detection datasets, and comparable performance on car viewpo- nt estimation.

Description: Convolutional Neural Network (CNN) has demonstrated promising performance in single-label image classification tasks. However, how CNN best copes with multi-label images still remains an open problem, mainly due to the complex underlying object layouts and insufficient multi-label training images. In this work, we propose a flexible deep CNN infrastructure, called Hypotheses-CNN-Pooling (HCP), where an arbitrary number of object segment hypotheses are taken as the inputs, then a shared CNN is connected with each hypothesis, and finally the CNN output results from different hypotheses are aggregated with max pooling to produce the ultimate multi-label predictions. Some unique characteristics of this flexible deep CNN infrastructure include: 1) no ground-truth bounding box information is required for training; 2) the whole HCP infrastructure is robust to possibly noisy and/or redundant hypotheses; 3) the shared CNN is flexible and can be well pre-trained with a large-scale single-label image dataset, e.g., ImageNet; and 4) it may naturally output multi-label prediction results. Experimental results on Pascal VOC 2007 and VOC 2012 multi-label image datasets well demonstrate the superiority of the proposed HCP infrastructure over other state-of-the-arts. In particular, the mAP reaches 90.5% by HCP only and 93.2% after the fusion with our complementary result in  [12] based on hand-crafted features on the VOC 2012 dataset.

Description: Wiberg matrix factorization breaks a matrix $Y$ into low-rank factors $U$ and $V$ by solving for $V$ in closed form given $U$ , linearizing $V(U)$ about $U$ , and iteratively minimizing $||Y - UV(U)||_2$ with respect to $U$ only. This approach factors the matrix while effectively removing $V$ from the minimization. Recently Eriksson and van den Hengel extended this approach to $L_1$ , minimizing $||Y - UV(U)||_1$ . We generalize their approach beyond factorization to minimize $||Y - f(U, V)||_1$ for more general functions $f(U, V)$ that are nonlinear in each of two sets of variables. We demonstrate the idea with a practical Wiberg algorithm for $L_1$ bundle adjustment. One Wiberg minimization can be nested inside another, effectively removing two of three sets of variables from a minimization. We demonstrate this idea with a nested Wiberg algorithm for $L_1$ projective bundle adjustment, solving for camera matrices, points, and projective depths. Wiberg minimization also generalizes to handle nonlinear constraints, and we demonstrate this idea with Constrained Wiberg Minimization for Multiple Instance Learning (CWM-MIL), which removes one set of variable

Description: In this paper, we propose a novel subspace learning algorithm called Local Feature Discriminant Projection (LFDP) for supervised dimensionality reduction of local features. LFDP is able to efficiently seek a subspace to improve the discriminability of local features for classification. We make three novel contributions. First, the proposed LFDP is a general supervised subspace learning algorithm which provides an efficient way for dimensionality reduction of large-scale local feature descriptors. Second, we introduce the Differential Scatter Discriminant Criterion (DSDC) to the subspace learning of local feature descriptors which avoids the matrix singularity problem. Third, we propose a generalized orthogonalization method to impose on projections, leading to a more compact and less redundant subspace. Extensive experimental validation on three benchmark datasets including UIUC-Sports, Scene-15 and MIT Indoor demonstrates that the proposed LFDP outperforms other dimensionality reduction methods and achieves state-of-the-art performance for image classification.

Description: We propose a real-time method to accurately track the human head pose in the 3-dimensional (3D) world. Using a RGB-Depth camera, a face template is reconstructed by fitting a 3D morphable face model, and the head pose is determined by registering this user-specific face template to the input depth video.

Description: Euclidean statistics are often generalized to Riemannian manifolds by replacing straight-line interpolations with geodesic ones. While these Riemannian models are familiar-looking, they are restricted by the inflexibility of geodesics, and they rely on constructions which are optimal only in Euclidean domains. We consider extensions of Principal Component Analysis (PCA) to Riemannian manifolds. Classic Riemannian approaches seek a geodesic curve passing through the mean that optimizes a criteria of interest. The requirements that the solution both is geodesic and must pass through the mean tend to imply that the methods only work well when the manifold is mostly flat within the support of the generating distribution. We argue that instead of generalizing linear Euclidean models, it is more fruitful to generalize non-linear Euclidean models. Specifically, we extend the classic Principal Curves from Hastie & Stuetzle to data residing on a complete Riemannian manifold. We show that for elliptical distributions in the tangent of spaces of constant curvature, the standard principal geodesic is a principal curve. The proposed model is simple to compute and avoids many of the pitfalls of traditional geodesic approaches. We empirically demonstrate the effectiveness of the Riemannian principal curves on several manifolds and datasets.

Description: An end-to-end real-time text localization and recognition method is presented. Its real-time performance is achieved by posing the character detection and segmentation problem as an efficient sequential selection from the set of Extremal Regions. The ER detector is robust against blur, low contrast and illumination, color and texture variation. In the first stage, the probability of each ER being a character is estimated using features calculated by a novel algorithm in constant time and only ERs with locally maximal probability are selected for the second stage, where the classification accuracy is improved using computationally more expensive features. A highly efficient clustering algorithm then groups ERs into text lines and an OCR classifier trained on synthetic fonts is exploited to label character regions. The most probable character sequence is selected in the last stage when the context of each character is known. The method was evaluated on three public datasets. On the ICDAR 2013 dataset the method achieves state-of-the-art results in text localization; on the more challenging SVT dataset, the proposed method significantly outperforms the state-of-the-art methods and demonstrates that the proposed pipeline can incorporate additional prior knowledge about the detected text. The proposed method was exploited as the baseline in the ICDAR 2015 Robust Reading competition, where it compares favourably to the state-of-the art.

Description: These instructions give guidelines for preparing papers for this publication. Presents information for authors publishing in this journal.

Description: This paper addresses modelling data using the Watson distribution. The Watson distribution is one of the simplest distributions for analyzing axially symmetric data. This distribution has gained some attention in recent years due to its modeling capability. However, its Bayesian inference is fairly understudied due to difficulty in handling the normalization factor. Recent development of Markov chain Monte Carlo (MCMC) sampling methods can be applied for this purpose. However, these methods can be prohibitively slow for practical applications. A deterministic alternative is provided by variational methods that convert inference problems into optimization problems. In this paper, we present a variational inference for Watson mixture models. First, the variational framework is used to side-step the intractability arising from the coupling of latent states and parameters. Second, the variational free energy is further lower bounded in order to avoid intractable moment computation. The proposed approach provides a lower bound on the log marginal likelihood and retains distributional information over all parameters. Moreover, we show that it can regulate its own complexity by pruning unnecessary mixture components while avoiding over-fitting. We discuss potential applications of the modeling with Watson distributions in the problem of blind source separation, and clustering gene expression data sets.

Description: Timing and carrier synchronization is a fundamental requirement for any wireless communication system to work properly. Timing synchronization is the process by which a receiver node determines the correct ins...

Description: Cognitive radio sensor networks (CRSNs) are multi-channel-capable networks that inherit some of the challenges of traditional wireless sensor networks (WSNs), such as limited power source and hardware capacity...

Description: In this paper, we study an infrastructure-less public safety network (IPSN) where energy efficiency and reliability are critical requirements in the absence of cellular infrastructure, i.e., base stations and ...

Description: An improved triple-pulse high-power pulse-width modulation (HP-PWM) signal model is proposed to enhance the performance of current underwater sonar transmitter. The presented model is based on reducing the thi...

Description: Sentiment analysis of online social media has attracted significant interest recently. Many studies have been performed, but most existing methods focus on either only textual content or only visual content. In this paper, we utilize deep learning models in a convolutional neural network (CNN) to analyze the sentiment in Chinese microblogs from both textual and visual content. We first train a CNN on top of pre-trained word vectors for textual sentiment analysis and employ a deep convolutional neural network (DNN) with generalized dropout for visual sentiment analysis. We then evaluate our sentiment prediction framework on a dataset collected from a famous Chinese social media network (Sina Weibo) that includes text and related images and demonstrate state-of-the-art results on this Chinese sentiment analysis benchmark.

Description: Stream classification service (SCS) is a novel concept proposed in the newly released IEEE 802.11aa standard for robust audio and video streaming, particularly for graceful degradation of streaming quality dur...

Description: We investigate the problem of minimizing the total power consumption under the constraint of the signal-to-noise ratio (SNR) requirement for the physical layer multicasting system with large-scale antenna arrays. In contrast with existing work, we explicitly consider both the transmit power and the circuit power scaling with the number of antennas. The joint antenna selection and beamforming technique is proposed to minimize the total power consumption. The problem is a challenging one, which aims to minimize the linear combination of ℓ 0 -norm and ℓ 2 -norm. To our best knowledge, this minimization problem has not yet been well solved. A random decremental antenna selection algorithm is designed, which is further modified by an approximation of the minimal transmit power based on the asymptotic orthogonality of the channels. Then, a more efficient decremental antenna selection algorithm is proposed based on minimizing the ℓ 0 norm. Performance results show that the ℓ 0 norm minimization algorithm greatly outperforms the random selection algorithm in terms of the total power consumption and the average run time.

Description: In order to improve the quality of multimedia communication and mobile users, the work efficiency of the cloud platform, the service-oriented mobile multimedia cooperative storing, and the delivery scheme were...

Description: Big data analytics (BDA) applications are a new category of software applications that process large amounts of data using scalable parallel processing infrastructure to obtain hidden value. Hadoop is the most...

Description: Owning to the high peak-to-average power ratio problem, the power efficiency of orthogonal-frequency-division-multiplexing (OFDM) systems is usually low. It deteriorates in millimeter wave systems in which the...

Description: Recently, head pose estimation (HPE) from low-resolution surveillance data has gained in importance. However, monocular and multi-view HPE approaches still work poorly under target motion , as facial appearance distorts owing to camera perspective and scale changes when a person moves around. To this end, we propose FEGA-MTL , a novel framework based on Multi-Task Learning (MTL) for classifying the head pose of a person who moves freely in an environment monitored by multiple, large field-of-view surveillance cameras. Upon partitioning the monitored scene into a dense uniform spatial grid, FEGA-MTL simultaneously clusters grid partitions into regions with similar facial appearance, while learning region-specific head pose classifiers. In the learning phase, guided by two graphs which a-priori model the similarity among (1) grid partitions based on camera geometry and (2) head pose classes, FEGA-MTL derives the optimal scene partitioning and associated pose classifiers. Upon determining the target's position using a person tracker at test time, the corresponding region-specific classifier is invoked for HPE. The FEGA-MTL framework naturally extends to a weakly supervised setting where the target's walking direction is employed as a proxy in lieu of head orientation. Experiments confirm that FEGA-MTL significantly outperforms competing single-task and multi-task learning methods in multi-view settings.

Description: Recent literature shows that facial attributes, i.e., contextual facial information, can be beneficial for improving the performance of real-world applications, such as face verification, face recognition, and image search. Examples of face attributes include gender, skin color, facial hair, etc. How to robustly obtain these facial attributes (traits) is still an open problem, especially in the presence of the challenges of real-world environments: non-uniform illumination conditions, arbitrary occlusions, motion blur and background clutter. What makes this problem even more difficult is the enormous variability presented by the same subject, due to arbitrary face scales, head poses, and facial expressions. In this paper, we focus on the problem of facial trait classification in real-world face videos. We have developed a fully automatic hierarchical and probabilistic framework that models the collective set of frame class distributions and feature spatial information over a video sequence. The experiments are conducted on a large real-world face video database that we have collected, labelled and made publicly available. The proposed method is flexible enough to be applied to any facial classification problem. Experiments on a large, real-world video database McGillFaces  [1] of 18,000 video frames reveal that the proposed framework outperforms alternative approaches, by up to 16.96 and 10.13%, for the facial attributes of gender and facial hair, respectively.