Advection‐Based Function Matching on Surfaces

A tangent vector field on a surface is the generator of a smooth family of maps from the surface to itself, known as the flow. Given a scalar function on the surface, it can be transported, or advected, by composing it with a vector field's flow. Such transport is exhibited by many physical phenomena, e.g., in fluid dynamics. In this paper, we are interested in the inverse problem: given source and target functions, compute a vector field whose flow advects the source to the target. We propose a method for addressing this problem, by minimizing an energy given by the advection constraint together with a regularizing term for the vector field. Our approach is inspired by a similar method in computational anatomy, known as LDDMM, yet leverages the recent framework of functional vector fields for discretizing the advection and the flow as operators on scalar functions. The latter allows us to efficiently generalize LDDMM to curved surfaces, without explicitly computing the flow lines of the vector field we are optimizing for. We show two approaches for the solution: using linear advection with multiple vector fields, and using non‐linear advection with a single vector field. We additionally derive an approximated gradient of the corresponding energy, which is based on a novel vector field transport operator. Finally, we demonstrate applications of our machinery to intrinsic symmetry analysis, function interpolation and map improvement.     

Applying Dynamic Training‐Subset Selection Methods Using Genetic Programming for Forecasting Implied Volatility

Volatility is a key variable in option pricing, trading, and hedging strategies. The purpose of this article is to improve the accuracy of forecasting implied volatility using an extension of genetic programming (GP) by means of dynamic training‐subset selection methods. These methods manipulate the training data in order to improve the out‐of‐sample patterns fitting. When applied with the static subset selection method using a single training data sample, GP could generate forecasting models, which are not adapted to some out‐of‐sample fitness cases. In order to improve the predictive accuracy of generated GP patterns, dynamic subset selection methods are introduced to the GP algorithm allowing a regular change of the training sample during evolution. Four dynamic training‐subset selection methods are proposed based on random, sequential, or adaptive subset selection. The latest approach uses an adaptive subset weight measuring the sample difficulty according to the fitness cases' errors. Using real data from S&P500 index options, these techniques are compared with the static subset selection method. Based on mean squared error total and percentage of non‐fitted observations, results show that the dynamic approach improves the forecasting performance of the generated GP models, especially those obtained from the adaptive‐random training‐subset selection method applied to the whole set of training samples.     

A modular system for global and local abnormal event detection and categorization in videos

This paper presents a modular system for both abnormal event detection and categorization in videos. Complementary normalcy models are built both globally at the image level and locally within pixels blocks. Three features are analyzed: (1) spatio-temporal evolution of binary motion where foreground pixels are detected using an enhanced background subtraction method that keeps track of temporarily static pixels; (2) optical flow, using a robust pyramidal KLT technique; and (3) motion temporal derivatives. At the local level, a normalcy MOG model is built for each block and for each flow feature and is made more compact using PCA. Then, the activity is analyzed qualitatively using a set of compact hybrid histograms embedding both optical flow orientation (or temporal gradient orientation) and foreground statistics. A compact binary signature of maximal size 13 bits is extracted from these different features for event characterization. The performance of the system is illustrated on different datasets of videos recorded on static cameras. The experiments show that the anomalies are well detected even if the method is not dedicated to one of the addressed scenarios.     

Evidential data mining: precise support and confidence

Associative classification has been shown to provide interesting results whenever of use to classify data. With the increasing complexity of new databases, retrieving valuable information and classifying incoming data is becoming a thriving and compelling issue. The evidential database is a new type of database that represents imprecision and uncertainty. In this respect, extracting pertinent information such as frequent patterns and association rules is of paramount importance task. In this work, we tackle the problem of pertinent information extraction from an evidential database. A new data mining approach, denoted EDMA, is introduced that extracts frequent patterns overcoming the limits of pioneering works of the literature. A new classifier based on evidential association rules is thus introduced. The obtained association rules, as well as their respective confidence values, are studied and weighted with respect to their relevance. The proposed methods are thoroughly experimented on several synthetic evidential databases and showed performance improvement.     

Parsing human skeletons in an operating room

Multiple human pose estimation is an important yet challenging problem. In an operating room (OR) environment, the 3D body poses of surgeons and medical staff can provide important clues for surgical workflow analysis. For that purpose, we propose an algorithm for localizing and recovering body poses of multiple human in an OR environment under a multi-camera setup. Our model builds on 3D Pictorial Structures and 2D body part localization across all camera views, using convolutional neural networks (ConvNets). To evaluate our algorithm, we introduce a dataset captured in a real OR environment. Our dataset is unique, challenging and publicly available with annotated ground truths. Our proposed algorithm yields to promising pose estimation results on this dataset.     

Prediction of visual attention with deep CNN on artificially degraded videos for studies of attention of patients with Dementia

Multimedia Tools and Applications - Studies of visual attention of patients with Dementia such as Parkinson’s Disease Dementia and Alzheimer Disease is a promising way for non-invasive...     

Endmember variability in Spectral Mixture Analysis: A review

The composite nature of remotely sensed spectral information often masks diagnostic spectral features and hampers the detailed identification and mapping of targeted constituents of the earth's surface. Spectral Mixture Analysis (SMA) is a well established and effective technique to address this mixture problem. SMA models a mixed spectrum as a linear or nonlinear combination of its constituent spectral components or spectral endmembers weighted by their subpixel fractional cover. By model inversion SMA provides subpixel endmember fractions. The lack of ability to account for temporal and spatial variability between and among endmembers has been acknowledged as a major shortcoming of conventional SMA approaches using a linear mixture model with fixed endmembers. Over the past decades numerous efforts have been made to circumvent this issue. This review paper summarizes the available methods and results of endmember variability reduction in SMA. Five basic principles to mitigate endmember variability are identified: (i) the use of multiple endmembers for each component in an iterative mixture analysis cycle, (ii) the selection of a subset of stable spectral features, (iii) the spectral weighting of bands, (iv) spectral signal transformations and (v) the use of radiative transfer models in a mixture analysis. We draw attention to the high complementarities between the different techniques and suggest that an integrated approach is necessary to effectively address endmember variability issues in SMA.     

A quantitative model for software engineering trends

Many decision-makers in industry, government and academia routinely make decisions whose outcome depends on the evolution of software technology trends. Even though the stakes of these decisions are usually very high, decision makers routinely depend on expert opinions and qualitative assessments to model the evolution of software technology; both of these sources of decision-making information are subjective, are based on opinions rather than facts, and are prone to error. In this paper, we report on our ongoing work to build quantitative models of the evolution of software technology trends. In particular, we discuss how we took specific evolutionary models and merged them into a single (general-purpose) model. The original specific models are derived empirically using statistical methods on trend data we had collected over several years, and have been validated individually; in this paper we further validate the generic (general-purpose) model.     

Projected sequential Gaussian processes: A C++ tool for interpolation of large datasets with heterogeneous noise

Heterogeneous datasets arise naturally in most applications due to the use of a variety of sensors and measuring platforms. Such datasets can be heterogeneous in terms of the error characteristics and sensor models. Treating such data is most naturally accomplished using a Bayesian or model-based geostatistical approach; however, such methods generally scale rather badly with the size of dataset, and require computationally expensive Monte Carlo based inference. Recently within the machine learning and spatial statistics communities many papers have explored the potential of reduced rank representations of the covariance matrix, often referred to as projected or fixed rank approaches. In such methods the covariance function of the posterior process is represented by a reduced rank approximation which is chosen such that there is minimal information loss. In this paper a sequential Bayesian framework for inference in such projected processes is presented. The observations are considered one at a time which avoids the need for high dimensional integrals typically required in a Bayesian approach. A C++ library, gptk, which is part of the INTAMAP web service, is introduced which implements projected, sequential estimation and adds several novel features. In particular the library includes the ability to use a generic observation operator, or sensor model, to permit data fusion. It is also possible to cope with a range of observation error characteristics, including non-Gaussian observation errors. Inference for the covariance parameters is explored, including the impact of the projected process approximation on likelihood profiles. We illustrate the projected sequential method in application to synthetic and real datasets. Limitations and extensions are discussed.