Search

Composition detection and semantic line detection are important research topics in computer vision and play an important auxiliary role in the analysis of image esthetics. However, at present, few researchers have considered the internal relationship between these two related tasks for comprehensive research. In order to solve this problem, we propose a synchronous detection network of composition class and semantic lines based on cross-attention, which can realize the mutual supervision and guidance between composition class detection and semantic line detection, to improve the accuracy of each other’s detection. First, the pre-trained composition detection model and the pre-trained semantic line detection model as two teacher models to provide data labels of composition and semantic line information for the student model. Then, we train a student model with the help of the teacher model. The student model adopts the multi-task learning architecture by combining soft and hard parameter sharing, as we propose. At the same time, we develop a cross-attention module to ensure that both tasks get the help and supervision they need from each other. Experimental results show that our method can draw semantic lines while detecting composition classes, which increases the interpretability of composition class detection. Our composition detection accuracy reaches 92.57%, and for benchmark semantic lines, the accuracy of our AUC_A metric can reach 92.00%. [ABSTRACT FROM AUTHOR]

Linear approximation methods suffer from Gibbs oscillations when approximating functions with jumps. Essentially non oscillatory subcell-resolution (ENO-SR) is a local technique avoiding oscillations and with a full order of accuracy, but a loss of regularity of the approximant appears. The goal of this paper is to introduce a new approach having both properties of full accuracy and regularity. In order to obtain it, we propose a three-stage algorithm: first, the data is smoothed by subtracting an appropriate non-smooth data sequence; then a chosen high order linear approximation operator is applied to the smoothed data and finally, an approximation with the proper jump or corner (jump in the first order derivative) discontinuity structure is reinstated by correcting the smooth approximation with the non-smooth element used in the first stage. This new procedure can be applied as subdivision scheme to design curves and surfaces both in point-value and in cell-average contexts. Using the proposed algorithm, we are able to construct approximations with high precision, with high piecewise regularity, and without smearing nor oscillations in the presence of discontinuities. These are desired properties in real applications as computer aided design or car design, among others. [ABSTRACT FROM AUTHOR]

The best polynomial approximation and the Chebyshev approximation are both important in numerical analysis. In tradition, the best approximation is regarded as better than the Chebyshev approximation, because it is usually considered in the uniform norm. However, it is not always superior to the latter, as noticed by Trefethen (Math. Today 47:184–188, 2011) for the function f (x) = | x - 0.25 | . Recently, Wang (arxiv, 2106.03456, 2021) observed a similar phenomenon for a function with an algebraic singularity and proved it in theory. In this paper, we find that for functions with logarithmic endpoint singularities, the pointwise errors of the Chebyshev approximation are smaller than those of the best approximations of the same degree, except only at the narrow boundary layer. The pointwise error for the Chebyshev series truncated at the degree n is O (n - κ) ( κ = min { 2 γ + 1 , 2 δ + 1 } ), but increases as O (n - (κ - 1)) at the endpoint singularities. Theorems are given to explain this effect. [ABSTRACT FROM AUTHOR]

We consider new degrees of freedom for higher order differential forms on cubical meshes. The approach is inspired by the idea of Rapetti and Bossavit to define higher order Whitney forms and their degrees of freedom using small simplices. We show that higher order differential forms on cubical meshes can be defined analogously using small cubes and prove that these small cubes yield unisolvent degrees of freedom. Importantly, this approach is compatible with discrete exterior calculus and expands the framework to cover higher order methods on cubical meshes, complementing the earlier strategy based on simplices. [ABSTRACT FROM AUTHOR]

Digitized handwritten documents have been used for various purposes, including age detection, a crucial area of research in fields like forensic investigation and medical diagnosis. Automated age recognition is deemed to be a difficult task, due to the great degree of similarity and overlap across people's handwriting. Consequently, the efficiency of the classification system is determined by extracting pertinent features from handwritten documents. This research proposes a set of age-related features suggested by a graphologist to classify handwritten documents into two age groups: youth adult and mature adult. The extracted features are: slant irregularity (SI), pen pressure irregularity (PPI), text lines irregularity (TLI) and the percentage of black and white pixels (PWB). Support Vector Machines (SVM) and Neural Network (NN) classifiers have been used to train, validate and test the proposed approach using two different datasets: the FSHS and the Khatt datasets. When applied to the FSHS dtaset using SVM and NN approaches, the proposed method resulted in a classification accuracy of 71% and 63.5%, respectively. Meanwhile, when applied to the Khatt dataset, our method outperformed state-of-the-art methods with a classification accuracy of 65.2% and 67% utilising SVM and NN classifiers, respectively. These are the best rates available right now in this field. [ABSTRACT FROM AUTHOR]

Facial expression generation has always been an intriguing task for scientists and researchers all over the globe. In this context, we present our novel approach for generating videos of the six basic facial expressions. Starting from a single neutral facial image and a label indicating the desired facial expression, we aim to synthesize a video of the given identity performing the specified facial expression. Our approach, referred to as FEV-GAN (Facial Expression Video GAN), is based on Spatio-temporal Convolutional GANs, that are known to model both content and motion in the same network. Previous methods based on such a network have shown a good ability to generate coherent videos with smooth temporal evolution. However, they still suffer from low image quality and low identity preservation capability. In this work, we address this problem by using a generator composed of two image encoders. The first one is pre-trained for facial identity feature extraction and the second for spatial feature extraction. We have qualitatively and quantitatively evaluated our model on two international facial expression benchmark databases: MUG and Oulu-CASIA NIR&VIS. The experimental results analysis demonstrates the effectiveness of our approach in generating videos of the six basic facial expressions while preserving the input identity. The analysis also proves that the use of both identity and spatial features enhances the decoder ability to better preserve the identity and generate high-quality videos. The code and the pre-trained model will soon be made publicly available.

An inverse central ordering of the nodes is proposed for the Newton interpolation formula. This ordering may improve the stability for certain distributions of nodes. For equidistant nodes, an upper bound of the conditioning is provided. This bound is close to the bound of the conditioning in the Lagrange interpolation formula, whose conditioning is the lowest. This ordering is related to a pivoting strategy of a matrix elimination procedure called Neville elimination. The results are illustrated with examples. [ABSTRACT FROM AUTHOR]

We propose an efficient, deterministic algorithm for constructing exponentially convergent deep neural network (DNN) approximations of multivariate, analytic maps f : [ - 1 , 1 ] K → R . We address in particular networks with the rectified linear unit (ReLU) activation function. Similar results and proofs apply for many other popular activation functions. The algorithm is based on collocating f in deterministic families of grid points with small Lebesgue constants, and by a-priori (i.e., “offline”) emulation of a spectral basis with DNNs to prescribed fidelity. Assuming availability of N function values of a possibly corrupted, numerical approximation f ˘ of f in [ - 1 , 1 ] K and a bound on ‖ f - f ˘ ‖ L ∞ ([ - 1 , 1 ] K) , we provide an explicit, computational construction of a ReLU DNN which attains accuracy ε (depending on N and ‖ f - f ˘ ‖ L ∞ ([ - 1 , 1 ] K) ) uniformly, with respect to the inputs. For analytic maps f : [ - 1 , 1 ] K → R , we prove exponential convergence of expression and generalization errors of the constructed ReLU DNNs. Specifically, for every target accuracy ε ∈ (0 , 1) , there exists N depending also on f such that the error of the construction algorithm with N evaluations of f ˘ as input in the norm L ∞ ([ - 1 , 1 ] K ; R) is smaller than ε up to an additive data-corruption bound ‖ f - f ˘ ‖ L ∞ ([ - 1 , 1 ] K) multiplied with a factor growing slowly with 1 / ε and the number of non-zero DNN weights grows polylogarithmically with respect to 1 / ε . The algorithmic construction of the ReLU DNNs which will realize the approximations, is explicit and deterministic in terms of the function values of f ˘ in tensorized Clenshaw–Curtis grids in [ - 1 , 1 ] K . We illustrate the proposed methodology by a constructive algorithm for (offline) computations of posterior expectations in Bayesian PDE inversion. [ABSTRACT FROM AUTHOR]

We present an asymptotic analysis of adaptive methods for Lp approximation of functions f ∈ Cr([a, b]), where 1 ≤ p ≤ + ∞ . The methods rely on piecewise polynomial interpolation of degree r − 1 with adaptive strategy of selecting m subintervals. The optimal speed of convergence is in this case of order m−r and it is already achieved by the uniform (nonadaptive) subdivision of the initial interval; however, the asymptotic constant crucially depends on the chosen strategy. We derive asymptotically best adaptive strategies and show their applicability to automatic Lp approximation with a given accuracy ε. [ABSTRACT FROM AUTHOR]

In order to solve Prandtl-type equations we propose a collocation-quadrature method based on de la Vallée Poussin (briefly VP) filtered interpolation at Chebyshev nodes. Uniform convergence and stability are proved in a couple of Hölder-Zygmund spaces of locally continuous functions. With respect to classical methods based on Lagrange interpolation at the same collocation nodes, we succeed in reproducing the optimal convergence rates of the L2 case and cut off the typical log factor which seemed inevitable dealing with uniform norms. Such an improvement does not require a greater computational effort. In particular, we propose a fast algorithm based on the solution of a simple 2-bandwidth linear system and prove that, as its dimension tends to infinity, the sequence of the condition numbers (in any natural matrix norm) tends to a finite limit. [ABSTRACT FROM AUTHOR]

• New model for scene text VQA that jointly reasons about textual and visual modalities. • Attending on multi-modal features is better than attending separately to each modality. • Grid features from an object detection backbone proves to work well in this task. • The proposed model is up to 5 times faster than current state of the art models. • Combined with a standard VQA model reaches state of the art on mixed datasets. This paper presents a new model for the task of scene text visual question answering. In this task questions about a given image can only be answered by reading and understanding scene text. Current state of the art models for this task make use of a dual attention mechanism in which one attention module attends to visual features while the other attends to textual features. A possible issue with this is that it makes difficult for the model to reason jointly about both modalities. To fix this problem we propose a new model that is based on an single attention mechanism that attends to multi-modal features conditioned to the question. The output weights of this attention module over a grid of multi-modal spatial features are interpreted as the probability that a certain spatial location of the image contains the answer text to the given question. Our experiments demonstrate competitive performance in two standard datasets with a model that is × 5 faster than previous methods at inference time. Furthermore, we also provide a novel analysis of the ST-VQA dataset based on a human performance study. Supplementary material, code, and data is made available through this link. [ABSTRACT FROM AUTHOR]

• A mental stress state diagnostic for timely intervention from clinicians. • Deep neural network with hierarchical learning capabilities using IoT based biomarkers. • Learns using both wrist-based and chest-based sensor bio-signal features. • Model-level fusion of high-level representations for stress state classification. Affective state recognition at an early stage can help in mood stabilization, stress and depression management for mental well-being. Pro-active and remote mental healthcare warrants the use of various biomarkers to detect the affective mental state of the individual by evaluating the daily activities. With the easy accessibility of IoT-based sensors for healthcare, observable and quantifiable characteristics of our body, physiological changes in the body can be measured and tracked using various wearable devices. This work puts forward a model for mental stress state detection using sensor-based bio-signals. A multi-level deep neural network with hierarchical learning capabilities of convolution neural network is proposed. Multivariate time-series data consisting of both wrist-based and chest-based sensor bio-signals is trained using a hierarchy of networks to generate high-level features for each bio-signal feature. A model-level fusion strategy is proposed to combine the high-level features into one unified representation and classify the stress states into three categories as baseline, stress and amusement. The model is evaluated on the WESAD benchmark dataset for mental health and compares favourably to state-of-the-art approaches giving a superlative performance accuracy of 87.7%. [ABSTRACT FROM AUTHOR]

This document summarises the current theoretical and experimental status of the di-Higgs boson production searches, and of the direct and indirect constraints on the Higgs boson self-coupling, with the wish to serve as a useful guide for the next years. The document discusses the theoretical status, including state-of-the-art predictions for di-Higgs cross sections, developments on the effective field theory approach, and studies on specific new physics scenarios that can show up in the di-Higgs final state. The status of di-Higgs searches and the direct and indirect constraints on the Higgs self-coupling at the LHC are presented, with an overview of the relevant experimental techniques, and covering all the variety of relevant signatures. Finally, the capabilities of future colliders in determining the Higgs self-coupling are addressed, comparing the projected precision that can be obtained in such facilities. The work has started as the proceedings of the Di-Higgs workshop at Colliders, held at Fermilab from the 4th to the 9th of September 2018, but it went beyond the topics discussed at that workshop and included further developments. FERMILAB-CONF-19-468-E-T, LHCHXSWG-2019-005

In 2012 the ATLAS and CMS collaborations discovered at the LHC the Higgs boson decaying to vector bosons. This discovery has provided a strong indication that the mechanism of Electroweak Symmetry Breaking (EWSB) is similar to the one predicted by Brout-Englert-Higgs (BEH) nearly 50 years before. Since then, one of the priorities of the LHC program, as well as of the majority of the future collider proposals, is to measure directly the parameters of the EWSB potential. The goal is to identify if it has indeed the straightforward quartic shape predicted by BEH or it is more complex, as the result of an unexplored physics nature. The answer to this major scientific question will have a considerable impact on our understanding of vacuum properties and the history of the universe through the EWSB during the Big Bang. The only direct way to probe these couplings is through the measure of the production of multiple Higgs bosons, two being the simplest case. In this paper, we present a comprehensive review of the current searches and the state of the art insights on the topic. In particular, we explain why this ambitious project is even more challenging than the discovery of the Higgs boson itself. Finally, we sketch the plans of the HEP community for how to access the parameters of the BEH mechanism. This review is adapted to a curious reader familiar with particle physics in general or a scientist who wants to have a landscape overview of the topic.

Imaging methods based on the absorption or scattering of atmospheric muons, collectively named under the neologism “muography”, exploit the abundant natural flux of muons produced from cosmic-ray interactions in the atmosphere. Recent years have seen a steep rise in the development of muography methods in a variety of innovative multidisciplinary approaches to study the interior of natural or human-made structures, establishing synergies between usually disconnected academic disciplines such as particle physics, geology, and archaeology. Muography also bears promise of immediate societal impact through geotechnical investigations, nuclear waste surveys, homeland security, and natural hazard monitoring. Our aim is to provide an introduction to this vibrant research area, starting from the physical principles at the basis of the methods and describing the main detector technologies and imaging tools, including their combination with conventional techniques from other disciplines, where appropriate. Then, we discuss critically some outstanding issues that affect a broad variety of applications, and the current state of the art in addressing them. Finally, we review several recent developments in the application of muography methods to specific use cases, without any pretence of exhaustiveness.

Demonstration of the light radiation pressure on a microscopic level by A. Ashkin led to the invention of optical tweezers (OT). Applied in the studies of living systems, OT have become a preferable instrument for the noninvasive study of microobjects, allowing manipulation and measurement of the mechanical properties of molecules, organelles, and cells. In the present paper, we overview OT applications in hemorheological research, placing emphasis on red blood cells but also discussing OT applications for the investigation of the biomechanics of leukocytes and platelets. Blood properties have always served as a primary parameter in medical diagnostics due to the interconnection with the physiological state of an organism. Despite blood testing being a well-established procedure of conventional medicine, there are still many complex processes that must be unraveled to improve our understanding and contribute to future medicine. OT are advancing single-cell research, promising new insights into individual cell characteristics compared to the traditional approaches. We review the fundamental and practical findings revealed in blood research through the optical manipulation, stretching, guiding, immobilization, and inter-/intracellular force measurements of single blood cells.

The complexity of collider data analyses has dramatically increased from early colliders to the CERN LHC. Reconstruction of the collision products in the particle detectors has reached a point that requires dedicated publications documenting the techniques, and periodic retuning of the algorithms themselves. Analysis methods evolved to account for the increased complexity of the combination of particles required in each collision event (final states) and for the need of squeezing every last bit of sensitivity from the data; physicists often seek to fully reconstruct the final state, a process that is mostly relatively easy at lepton colliders but sometimes exceedingly difficult at hadron colliders to the point of requiring sometimes using advanced statistical techniques such as machine learning. The need for keeping the publications documenting results to a reasonable size implies a greater level of compression or even omission of information with respect to publications from twenty years ago. The need for compression should however not prevent sharing a reasonable amount of information that is essential to understanding a given analysis. Infrastructures like Rivet or HepData have been developed to host additional material, but physicists in the experimental Collaborations often still send an insufficient amount of material to these databases. In this manuscript I advocate for an increase in the information shared by the Collaborations, and try to define a minimum standard for acceptable level of information when reporting the results of statistical procedures in High Energy Physics publications.

• Enhance face verification performance for demographics effected by decision bias. • Operates on comparison score-level to enable using existing face recognition models. • Based on the notation of individual fairness to treat similar face groups similarly. • Effectiveness proved on three databases and two face embeddings. • Unsupervised enhancement of the overall and intra-class recognition performance. Current face recognition systems achieve high progress on several benchmark tests. Despite this progress, recent works showed that these systems are strongly biased against demographic sub-groups. Consequently, an easily integrable solution is needed to reduce the discriminatory effect of these biased systems. Previous work mainly focused on learning less biased face representations, which comes at the cost of a strongly degraded overall recognition performance. In this work, we propose a novel unsupervised fair score normalization approach that is specifically designed to reduce the effect of bias in face recognition and subsequently lead to a significant overall performance boost. Our hypothesis is built on the notation of individual fairness by designing a normalization approach that leads to treating "similar" individuals "similarly". Experiments were conducted on three publicly available datasets captured under controlled and in-the-wild circumstances. Results demonstrate that our solution reduces demographic biases, e.g. by up to 82.7% in the case when gender is considered. Moreover, it mitigates the bias more consistently than existing works. In contrast to previous works, our fair normalization approach enhances the overall performance by up to 53.2% at false match rate of 10 − 3 and up to 82.9% at a false match rate of 10 − 5. Additionally, it is easily integrable into existing recognition systems and not limited to face biometrics. [ABSTRACT FROM AUTHOR]

In this paper, the properties of a new family of nonlinear dyadic subdivision schemes are presented and studied depending on the conditions imposed to the mean used to rewrite the linear scheme upon which the new scheme is based. The convergence, stability, and order of approximation of the schemes of the family are analyzed in general. Also, the elimination of the Gibbs oscillations close to discontinuities is proved in particular cases. It is proved that these schemes converge towards limit functions that are Hölder continuous with exponent larger than 1. The results are illustrated with several examples. [ABSTRACT FROM AUTHOR]

This paper is devoted to the construction and analysis of some new non-linear subdivision and multiresolution schemes using the Lehmer means. Our main objective is to attain adaption close to discontinuities. We present theoretical, numerical results and applications for different schemes. The main theoretical result is related to the four-point interpolatory scheme, that we write as a perturbation of a linear scheme. Our aim is to establish a one-step contraction property that allows to prove the stability of the new scheme. Indeed with a one-step contraction property for the scheme of differences, it is possible to prove the stability of the 2D algorithm constructed using a tensor product approach. In this article, we also consider the associated three points cell-average scheme, that we will use to present some results for image compression, and a non-interpolatory scheme, that we will use to introduce an application to subdivision curves in 2D. These applications show that the use of the Lehmer mean is suitable for the design of subdivision schemes for the generation of curves and for image processing. [ABSTRACT FROM AUTHOR]

A geometric construction of a Bézier curve is presented by a unifiable way from the mentioned literature with some modification. A closed-form solution to the inverse problem in cubic Bézier-spline interpolation will be obtained. Calculations in the given examples are performed by a Maple procedure using this solution. [ABSTRACT FROM AUTHOR]

• We propose a module to predict boundary shapes and boxes with a new masking scheme. • We comprehensively investigate proposed frameworks with various masking methods. • The proposed modularizable component can easily be applied to existing models. • BshapeNet + improves accuracy notably compared to the baselines, Faster and Mask RCNN. • We achieved highly competitive results with SOTA detection and segmentation models. We propose a modularizable component that can predict the boundary shapes and boxes of an image, along with a new masking scheme for improving object detection and instance segmentation. Specifically, we introduce two types of novel masks: a bounding box (bbox) mask and a bounding shape (bshape) mask. For each of these types, we consider two variants—the "Thick" model and the "Scored" model—both of which have the same morphology but differ in ways that make their boundaries thicker. To evaluate our masks, we design extended frameworks by adding a bshape mask (or a bbox mask) branch to a Faster R-CNN, and call this BshapeNet (or BboxNet). Furthermore, we propose BshapeNet+, a network that combines a bshape mask branch with a Mask R-CNN. Among our various models, BshapeNet+ demonstrates the best performance in both tasks. In addition, BshapeNet+ markedly outperforms the baseline models on MS COCO and Cityscapes and achieves highly competitive results with state-of-the-art models. In particular, the experimental results show that our branch works well on small objects and is easily applicable to various models, such as PANet as well as Faster R-CNN and Mask R-CNN. [ABSTRACT FROM AUTHOR]

The stability properties of the Newton interpolation formula depend on the order of the nodes and can be measured through a condition number. Increasing and Leja orderings have been previously considered (Carnicer et al. in J Approx Theory, 2017. https://doi.org/10.1016/j.jat.2017.07.005; Reichel in BIT 30:332–346, 1990). We analyze central orderings for equidistant nodes on a bounded real interval. A bound for conditioning is given. We demonstrate in particular that this ordering provides a more stable Newton formula than the natural increasing order. We also analyze of a central ordering with respect to the evaluation point, which provides low bounds for the conditioning. Numerical examples are included. [ABSTRACT FROM AUTHOR]

This paper examines the problem of extrapolation of an analytic function for x > 1 given N + 1 perturbed samples from an equally spaced grid on [ - 1 , 1 ] . For a function f on [ - 1 , 1 ] that is analytic in a Bernstein ellipse with parameter ρ > 1 , and for a uniform perturbation level ε on the function samples, we construct an asymptotically best extrapolant e(x) as a least squares polynomial approximant of degree M ∗ determined explicitly. We show that the extrapolant e(x) converges to f(x) pointwise in the interval I ρ ∈ [ 1 , (ρ + ρ - 1) / 2) as ε → 0 , at a rate given by a x-dependent fractional power of ε . More precisely, for each x ∈ I ρ we have | f (x) - e (x) | = O ε - log r (x) / log ρ , r (x) = x + x 2 - 1 ρ , up to log factors, provided that an oversampling conditioning is satisfied, viz. M ∗ ≤ 1 2 N , which is known to be needed from approximation theory. In short, extrapolation enjoys a weak form of stability, up to a fraction of the characteristic smoothness length. The number of function samples does not bear on the size of the extrapolation error provided that it obeys the oversampling condition. We also show that one cannot construct an asymptotically more accurate extrapolant from equally spaced samples than e(x), using any other linear or nonlinear procedure. The proofs involve original statements on the stability of polynomial approximation in the Chebyshev basis from equally spaced samples and these are expected to be of independent interest. [ABSTRACT FROM AUTHOR]

Highlights • We propose trainable feature extractors for representation of skeleton poses. • We employ the proposed feature extractor for classification of human actions. • We designed an human action classification method based on string kernel. • We carried out experiments on the MHAD, MSRDA and MIVIA-S action datasets. • We publicly released the MIVIA-S dataset for research purpose. Abstract Automatic interpretation of human actions gained strong interest among researchers in patter recognition and computer vision because of its wide range of applications, such as in social and home robotics, elderly people health care, surveillance, among others. In this paper, we propose a method for recognition of human actions by analysis of skeleton poses. The method that we propose is based on novel trainable feature extractors, which can learn the representation of prototype skeleton examples and can be employed to recognize skeleton poses of interest. We combine the proposed feature extractors with an approach for classification of pose sequences based on string kernels. We carried out experiments on three benchmark data sets (MIVIA-S, MSRSDA and MHAD) and the results that we achieved are comparable or higher than the ones obtained by other existing methods. A further important contribution of this work is the MIVIA-S dataset, that we collected and made publicly available. [ABSTRACT FROM AUTHOR]

We discuss matrix polynomials expressed in a Newton basis, and the associated polynomial eigenvalue problems. Properties of the generalized ansatz spaces associated with such polynomials are proved directly by utilizing a novel representation of pencils in these spaces. Also, we show how the family of Fiedler pencils can be adapted to matrix polynomials expressed in a Newton basis. These new Newton–Fiedler pencils are shown to be strong linearizations, and some computational aspects related to them are discussed. [ABSTRACT FROM AUTHOR]

Multi-output least-squares support vector regression machines (MLS-SVR) is proposed by Xu et al. [29] to handle multi-output regression problems. However, the prohibitive cost of model selection severely hinders MLS-SVR’s application. In this paper, an efficient gradient-based model selection algorithm for MLS-SVR is proposed. Firstly, a new training algorithm for MLS-SVR is developed, which allows one to obtain the solution vector for each output independently by dealing with matrices of much lower order. Based on the new training algorithm, a new leave-one-out error estimate is derived through virtual leave-one-out cross-validation. The model selection criterion is based on the new leave-one-out error estimate and its derivatives with respect to the hyper-parameters are also derived analytically. Both the model selection criterion and its partial derivatives can be obtained straightway once a training process ended. Finally, the hyper-parameters corresponding to the lowest model selection criterion is obtained through gradient decent method. The effectiveness and generalization performance of the proposed algorithm are validated through experiments on several multi-output datasets. Experiment results show that the proposed algorithm can save computational time dramatically without losing accuracy. [ABSTRACT FROM AUTHOR]

We propose a novel single face image super-resolution method, which is named Face Conditional Generative Adversarial Network (FCGAN), based on boundary equilibrium generative adversarial networks. Without taking any prior facial information, our approach combines the pixel-wise L 1 loss and GAN loss to optimize our super-resolution model and to generate a high-quality face image from a low-resolution one robustly (with upscaling factor 4 × ). Additionally, Compared with existing peer researches, both training and testing phases of FCGAN are end-to-end pipeline without pre/post-processing. To enhance the convergence speed and strengthen feature propagation, the Generator and Discriminator networks are designed with a skip-connection architecture, and both using an auto-encoder structure. Quantitative experiments demonstrate that our model achieves competitive performance compared with the state-of-the-art models based on both visual quality and quantitative criterions. We believe this high-quality face image generated method can impact many applications in face identification and intelligent monitor. [ABSTRACT FROM AUTHOR]

Recently convolutional neural networks (CNN) have demonstrated remarkable performance in various classification problems. In this paper, we also introduce CNN into in-air handwritten Chinese character recognition (IAHCCR) and propose new directional feature maps, named bend directional feature maps. Then we integrate the combination of various types of directional feature maps with the CNN and obtain better recognition performance compared with other methods reported for IAHCCR. For further improving recognition rate, we propose a new data augmentation method dedicated to in-air handwritten Chinese characters. The proposed data augmentation method combines global transformation with local distortion and effectively enlarges the training dataset. Experimental results demonstrate that our proposed methods can greatly improve the recognition rate for IAHCCR. [ABSTRACT FROM AUTHOR]

Decision Trees and Random Forests are among the most popular methods for classification tasks. Two key issues faced by these methods are: how to select the best attribute to associate with a node and how to split the samples given the selected attribute. This paper addresses an important challenge that arises when nominal attributes with a large number of values are present: the computational time required to compute splits of good quality. We present a framework to generate computationally efficient splitting criteria that handle, with theoretical approximation guarantee, multi-valued nominal attributes for classification tasks with a large number of classes. Experiments with a number of datasets suggest that a method derived from our framework is competitive in terms of accuracy and speed with the Twoing criterion, one of few criteria available that is able to handle, with optimality guarantee, nominal attributes with a large number of distinct values. However, this method has the advantage of also efficiently handling datasets with a large number of classes. These experiments also give evidence of the potential of aggregating attributes to improve the classification power of decision trees. [ABSTRACT FROM AUTHOR]

Identification of prescription patterns is a useful and interesting goal from multiple perspectives. The identified prescription patterns may expand the horizons of medical knowledge, and may be evaluated by subject matter experts to label certain patterns as anomalies calling for further investigation; for example, in prescription costs for insurance companies. This paper presents the statistical modeling details of the tag hierarchical topic model (Tag-HTM) and its application to the Health Insurance Review & Assessment Service (HIRA) dataset. The implementation of Tag-HTM revealed a hierarchical structure for medicine symptom distributions, which could constitute a new hierarchical categorization for diseases. The experimental results demonstrate that our generated hierarchical structure can replicate the existing hierarchy, namely ICD-10, which has been created by medical subject matter experts, to a considerable extent. Furthermore, the experiments indicate a quantitative performance improvement; that is, the superior perplexities of Tag-HTM compared to baselines. Moreover, Tag-HTM was able to isolate prescription patterns with higher medical costs as a branch of hierarchical clustering, and this cluster could form a prescription collection of interest to subject matter experts in insurance companies. [ABSTRACT FROM AUTHOR]

In recent years, QR code has been widely used in various society fields, and has provided great convenience in information exchange, commodity payment and website registration. However, the degradation of QR code images can cause difficulty in reading the code information. In this paper, we propose an alternating minimization restoration model for QR code images. Based on the binary prior, we use the binary characteristic and L 0 norm as the regularization terms, and introduce different auxiliary variables to make the complex model become solvable. The bi-level constraint of bar code image maintains the sparseness of image pixels and gradient. The experimental result shows that our algorithm has excellent effect on common blur kernels and multiple noise, and its performance is better than the state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Multi set canonical correlation analysis (mCCA), which extends the application of canonical correlation analysis (CCA) to more than two datasets, is a data driven technique that can jointly analyze the relationship amongst multiple (more than two) datasets. However, the conventional mCCA is directly applicable only to multivariate vector data and requires the image data to be reshaped into vectors. This approach fails to consider the spatial structure of the images and in addition, leads to an increase in the computational complexity. In this paper, we propose new two dimensional mCCA algorithms that operate directly on the image data instead of vectorizing them. Face recognition experiments are presented to compare the performances of conventional mCCA and the proposed two dimensional mCCA techniques. Additionally, experiments against fMRI data are conducted to demonstrate the applicability of the proposed approach in multisubject fMRI analysis. [ABSTRACT FROM AUTHOR]

Many modern applications deal with multi-label data, such as functional categorizations of genes, image labeling and text categorization. Classification of such data with a large number of labels and latent dependencies among them is a challenging task, and it becomes even more challenging when the data is received online and in chunks. Many of the current multi-label classification methods require a lot of time and memory, which make them infeasible for practical real-world applications. In this paper, we propose a fast linear label space dimension reduction method that transforms the labels into a reduced encoded space and trains models on the obtained pseudo labels. Additionally, it provides an analytical method to update the decoding matrix which maps the labels into the original space and is used during the test phase. Experimental results show the effectiveness of this approach in terms of running times and the prediction performance over different measures. [ABSTRACT FROM AUTHOR]

This work proposes a distance that combines Minkowski and Chebyshev distances and can be seen as an intermediary distance. This combination not only achieves efficient run times in neighbourhood iteration tasks in Z 2 , but also obtains good accuracies when coupled with the k-Nearest Neighbours (k-NN) classifier. The proposed distance is approximately 1.3 times faster than Manhattan distance and 329.5 times faster than Euclidean distance in discrete neighbourhood iterations. An accuracy analysis of the k-NN classifier using a total of 33 datasets from the UCI repository, 15 distances and values assigned to k that vary from 1 to 200 is presented. In this experiment, the proposed distance obtained accuracies that were better than the average more often than its counterparts (in 26 cases out of 33), and also obtained the best accuracy more frequently (in 9 out of 33 cases). [ABSTRACT FROM AUTHOR]

Suicide by hanging is a sentinel event and a major cause of death in prisons, with an increasing frequency over recent years. The rapid detection of suicidal behavior can reduce the mortality rate and increase the odds of survival for the suicide victim. Significant efforts have been made to develop technologies for preventing hanging attempts, but most of them use cumbersome devices, or they are mainly depending on human attention and intervention. In this paper, we propose a vision-based method to automatically detect suicide by hanging. Our intelligent video surveillance system operates using depth streams provided by an RGB-D camera, regardless of illumination conditions. The proposed algorithm is based on the exploitation of the body joints’positions to model suicidal behavior. Both dynamic and static pose characteristics are calculated in order to efficiently capture the body joints’movement and model suicidal behavior. Results from the experiments on realistic video sequences, show that our system achieves a high accuracy in detecting suicide attempts, while meeting real-time requirements. [ABSTRACT FROM AUTHOR]

Video prediction plays a fundamental role in video analysis and pattern recognition. However, the generated future frames are often blurred, which are not sufficient for further research. To overcome this obstacle, this paper proposes a new deep generative video prediction network under the framework of generative adversarial nets. The network consists of three components: a motion encoder, a frame generator and a frame discriminator. The motion encoder receives multiple frame differences (also known as Eulerian motion ) as input and outputs a global video motion representation. The frame generator is a pseudo-reverse two-stream network to generate the future frame. The frame discriminator is a discriminative 3D convolution network to determine whether the given frame is derived from the true future frame distribution or not. The frame generator and frame discriminator train jointly in an adversarial manner until a Nash equilibrium. Motivated by theories on color filter array, this paper also designs a novel cross channel color gradient (3CG) loss as a guidance of deblurring. Experiments on two state-of-the-art data sets demonstrate that the proposed network is promising. [ABSTRACT FROM AUTHOR]

It is generally much harder to reconstruct the 3D structures of a non-rigid object than that of a rigid object, because of the increased degrees of freedom. To overcome the ill-posed nature of this problem, there have been many researches that incorporate prior assumptions on the object. An attractive approach is to assume the local parts to be rigid, because there might be a possibility to reduce some degrees of freedom that have to be estimated. Unfortunately, this assumption is not strictly true for general objects, and rigid and non-rigid parts are usually mixed in an object. This issue may be circumvented if we can utilize a robust local-rigidity prior to enforce rigidity on local parts selectively. In this paper, we provide a simple cost function that imposes a rigidity prior on each local triangular patch, and the proposed algorithm minimizes this cost to reconstruct a non-rigid surface. Non-rigid local parts are handled as outliers in the proposed algorithm, which makes the algorithm capable of reconstructing a near-locally-rigid surface that might contain a few non-negligible deformations. Experimental results show that the proposed method provides the state-of-the-art performance for well-known dense surface data sets. [ABSTRACT FROM AUTHOR]

Previous Spectral Feature Selection (SFS) methods output promising feature selection results in many real-world applications, which deeply depend on the preservation of the local or global structures of the data via learning a graph matrix. However, current SFS methods 1) learn the graph matrix in the original data which may contain a number of noise to affect the results of feature selection, 2) conduct the learning of a low-dimensional feature space and the graph matrix individually, thus hard achieve the optimal results of feature selection even though both of these two steps achieve their individual optimization, and 3) consider either the local or global structure of data to difficult provide complementary information for feature selection. To address the above issues, this paper proposes a novel supervised feature selection algorithm to simultaneously preserve the local structure (via adaptive structure learning in a low-dimensional feature space of the original data) and the global structure (via a low-rank constraint) of the data. Moreover, we also propose a new optimization method to fast optimize the resulting objective function. We finally verify the proposed method on eight real-word and benchmark datasets, by comparing with the state-of-the-art feature selection methods, and experimental results show that our proposed method achieves competitive results in term of classification performance. [ABSTRACT FROM AUTHOR]

So far, most of existing feature selection methods have two defects: 1) These methods are more or less heavy workload, 2) Their effect is not very good enough. To solve the above issues, a novel unsupervised feature selection algorithm is proposed. Specifically, the proposed method uses the property of the data to construct self-representation coefficient matrix, and utilizes sparse representation to find the sparse structure of the self-representation coefficient matrix, and embeds a hypergraph Laplacian regularization term to make up the insignificance of ordinary graph in the representation of multiple relations. The Linear Discriminant Analysis (LDA) algorithm is used to further adjust the result of feature selection. Finally, a low rank constraint is used to capture global structure of data. Experimental results on real datasets showed that the proposed method outperformed the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

This paper studies an example-driven k -parameter computation that identifies different k values for different test samples in k NN prediction applications, such as classification, regression and missing data imputation. This is carried out with reconstructing a sparse coefficient matrix between test samples and training data. In the reconstruction process, an ℓ 1 − norm regularization is employed to generate an element-wise sparsity coefficient matrix, and an LPP (Locality Preserving Projection) regularization is adopted to keep the local structures of data for achieving the efficiency. Further, with the learnt k value, k NN approach is applied to classification, regression and missing data imputation. We experimentally evaluate the proposed approach with 20 real datasets, and show that our algorithm is much better than previous k NN algorithms in terms of data mining tasks, such as classification, regression and missing value imputation. [ABSTRACT FROM AUTHOR]

Load forecasting is an important part of power grid management. Accurate and timely load forecasting is of great significance to formulate economical and reasonable power allocation plan, improve safety and economy of power grid operation and improve power quality. In this paper, in order to find electricity load forecasting model, we propose an electricity load forecasting function mining algorithm based on artificial fish swarm and gene expression programming (ELFFM-AFSGEP). On the basis, distributed load forecast model mining based on hybrid gene expression programming and cloud computing (DLFMM-HGEPCloud) is proposed to solve the problem of massive electricity load forecasting. In order to better solve global electricity load forecasting model, error minimization crossover is introduced into DLFMM-HGEPCloud. The performance of the proposed algorithm in this paper is evaluated with a real-world dataset, and compared with GEP and some published algorithms by using the same dataset. Experimental results show that our proposed algorithm has an advantage in average time-consumption, average number of convergence, forecasted accuracy and excellent parallel performance in speedup and scaleup. [ABSTRACT FROM AUTHOR]