Bayesian Covariance Representation with Global Informative Prior for 3D Action Recognition

For the merits of high-order statistics and Riemannian geometry, covariance matrix has become a generic feature representation for action recognition. An independent action can be represented by an empirical statistics over all of its pose samples. Two major problems of covariance include the following: (1) it is prone to be singular so that actions fail to be represented properly, and (2) it is short of global action/pose-aware information so that expressive and discriminative power is limited. In this article, we propose a novel Bayesian covariance representation by a prior regularization method to solve the preceding problems. Specifically, covariance is viewed as a parametric maximum likelihood estimate of Gaussian distribution over local poses from an independent action. Then, a Global Informative Prior (GIP) is generated over global poses with sufficient statistics to regularize covariance. In this way, (1) singularity is greatly relieved due to sufficient statistics, (2) global pose information of GIP makes Bayesian covariance theoretically equivalent to a saliency weighting covariance over global action poses so that discriminative characteristics of actions can be represented more clearly. Experimental results show that our Bayesian covariance with GIP efficiently improves the performance of action recognition. In some databases, it outperforms the state-of-the-art variant methods that are based on kernels, temporal-order structures, and saliency weighting attentions, among others.

Underwater Target Perception in Local HOS Space

In this paper, we propose an underwater target perception architecture, which adopts the three-stage processing including underwater scene acoustic imaging, local high-order statistics (HOS) space conversion, and region-of-interest (ROI) detection. After analysing the problem of the underwater targets represented by the acoustic images, the unique cube structure of the target in local skewness space is noticed, which is used as a clue to develop the ROI detection of underwater scenes. In order to restore the actual appearance of the ROI as much as possible, the focus processing is explored to achieve the target reconstruction. When the target size and number are unknown, using an uncertain theoretical template can achieve a better target reconstruction effect. The performance of the proposed method in terms of SNR, detection rate, and false alarm rate is verified by experiments with several acoustic image sequences. Moreover, target perception architecture is general and can be generalized to a wider range of underwater applications.

Artifact Removing From Eeg Recordings Using Independent Component Analysis With High-order Statistics

Many methods have been proposed to remove artifacts from EEG recordings es- pecially those arising from eye movements and blinks. Often regression in time and frequency domain on parallel EEG and electrooculographic recordings is used, but this approach can become problematic in some cases. Use of Principal Com- ponent Analysis (PCA) has been proposed to re- move eye artifacts from multichannel EEG. This method is not effective when the activations from cerebral activity and artifacts have comparable amplitudes. In this paper it is presented a gener- ally applicable method for removing a wide vari- ety of artifacts from EEG recordings based on In- dependent Component Analysis (ICA) with high- order statistics. The method is applied with good results in the analysis of a sample lowpass event -related potentials (ERP) data.

Brain Function Network: Higher Order vs. More Discrimination

Brain functional network (BFN) has become an increasingly important tool to explore individual differences and identify neurological/mental diseases. For estimating a “good” BFN (with more discriminative information for example), researchers have developed various methods, in which the most popular and simplest is Pearson's correlation (PC). Despite its empirical effectiveness, PC only encodes the low-order (second-order) statistics between brain regions. To model high-order statistics, researchers recently proposed to estimate BFN by conducting two sequential PCs (denoted as PC2 in this paper), and found that PC2-based BFN can provide additional information for group difference analysis. This inspires us to think about (1) what will happen if continuing the correlation operation to construct much higher-order BFN by PCn (n>2), and (2) whether the higher-order correlation will result in stronger discriminative ability. To answer these questions, we use PCn-based BFNs to predict individual differences (Female vs. Male) as well as identify subjects with mild cognitive impairment (MCI) from healthy controls (HCs). Through experiments, we have the following findings: (1) with the increase of n, the discriminative ability of PCn-based BFNs tends to decrease; (2) fusing the PCn-based BFNs (n>1) with the PC1-based BFN can generally improve the sensitivity for MCI identification, but fail to help the classification accuracy. In addition, we empirically find that the sequence of BFN adjacency matrices estimated by PCn (n = 1,2,3,⋯ ) will converge to a binary matrix with elements of ± 1.

Feature Space Targeted Attacks by Statistic Alignment

By adding human-imperceptible perturbations to images, DNNs can be easily fooled. As one of the mainstream methods, feature space targeted attacks perturb images by modulating their intermediate feature maps, for the discrepancy between the intermediate source and target features is minimized. However, the current choice of pixel-wise Euclidean Distance to measure the discrepancy is questionable because it unreasonably imposes a spatial-consistency constraint on the source and target features. Intuitively, an image can be categorized as "cat'' no matter the cat is on the left or right of the image. To address this issue, we propose to measure this discrepancy using statistic alignment. Specifically, we design two novel approaches called Pair-wise Alignment Attack and Global-wise Alignment Attack, which attempt to measure similarities between feature maps by high-order statistics with translation invariance. Furthermore, we systematically analyze the layer-wise transferability with varied difficulties to obtain highly reliable attacks. Extensive experiments verify the effectiveness of our proposed method, and it outperforms the state-of-the-art algorithms by a large margin. Our code is publicly available at https://github.com/yaya-cheng/PAA-GAA.

The solar wind between 0.1 and 1 AU: Parker Solar Probe - BepiColombo radial alignment and BepiColombo - ACE magnetic alignment

<p>At the beginning of September 2020 ACE and BepiColombo spent several hours in an interesting magnetically connected configuration, while at the end of the same month Parker Solar Probe (PSP) and BepiColombo were radially aligned. Being PSP orbiting near 0.1 AU, BepiColombo near 0.6 AU, and ACE at 1 AU, these geometries are of particular interest for investigating the evolution of solar wind properties at different heliocentric distances by observing the same solar wind plasma parcels.<span class="Apple-converted-space"> </span></p> <p>In this contribution we use magnetic field observations from pairs of spacecraft to characterize both the topology of the magnetic field at different heliocentric distances (scalings and high-order statistics) and how it evolves when moving from near-Sun to far-Sun locations. We observe a breakdown of the statistical self-similar nature of the solar wind plasma due to an increase of the intermittency level when moving away from the Sun. These results support previous evidences on the radial dependence of solar wind scaling behavior and can open a novel framework for modeling magnetic field topological changes across the Heliosphere.</p>

High-Order Data-Driven Spatial Simulation of Categorical Variables

Modern approaches for the spatial simulation of categorical variables are largely based on multi-point statistical methods, where a training image is used to derive complex spatial relationships using relevant patterns. In these approaches, simulated realizations are driven by the training image utilized, while the spatial statistics of the actual sample data are ignored. This paper presents a data-driven, high-order simulation approach based on the approximation of high-order spatial indicator moments. The high-order spatial statistics are expressed as functions of spatial distances that are similar to variogram models for two-point methods, while higher-order statistics are connected with lower-orders via boundary conditions. Using an advanced recursive B-spline approximation algorithm, the high-order statistics are reconstructed from the available data and are subsequently used for the construction of conditional distributions using Bayes’ rule. Random values are subsequently simulated for all unsampled grid nodes. The main advantages of the proposed technique are its ability to (a) simulate without a training image to reproduce the high-order statistics of the data, and (b) adapt the model’s complexity to the information available in the data. The practical intricacies and effectiveness of the proposed approach are demonstrated through applications at two copper deposits.

SPA-Based Modified Local Reachability Density Ratio wSVDD for Nonlinear Multimode Process Monitoring

Many industrial processes are operated in multiple modes due to different manufacturing strategies. Multimodality of process data is often accompanied with nonlinear and non-Gaussian characteristics, which makes data-driven monitoring more complicated. In this paper, statistics pattern analysis (SPA) is introduced to extract low- and high-order statistics from raw process data. Support vector data description (SVDD), which can deal with nonlinear and non-Gaussian problems, is applied to monitor multimode process in this paper. To improve detection performance of SVDD for training multimode data with outliers, modified local reachability density ratio (mLRDR) is proposed as a weight factor to be embedded in the weighted-SVDD (wSVDD) model, in which the local neighbors in terms of both space and time are considered. Finally, the effectiveness and superiority of our proposed method are demonstrated by the Tennessee-Eastman (TE) process and wastewater treatment process (WWTP).

Artifact Removing From Eeg Recordings Using Independent Component Analysis with High-order Statistics

Many methods have been proposed to remove artifacts from EEG recordings especially those arising from eye movements and blinks. Often regression in time and frequency domain on parallel EEG and electrooculographic recordings is used, but this approach can become problematic in some cases. Use of Principal Component Analysis (PCA) has been proposed to remove eye artifacts from multichannel EEG. This method is not effective when the activations from cerebral activity and artifacts have comparable amplitudes. In this paper it is presented a generally applicable method for removing a wide variety of artifacts from EEG recordings based on Independent Component Analysis (ICA) with highorder statistics. The method is applied with good results in the analysis of a sample lowpass event -related potentials (ERP) data.