Long-distance Correlation Research Articles

Hurst Exponent and Event-by-Event Fluctuations in Relativistic Nucleus–Nucleus Collisions

A joint study of multi-particle pseudo-rapidity correlations and event-by-event fluctuations in the distributions of secondary particles and fragments of the target nucleus and the projectile nucleus was carried out in order to search for correlated clusters of secondary particles. An analysis of the collisions of the sulfur nucleus with photoemulsion nuclei at an energy of 200 A·GeV is presented based on experimental data obtained at the SPS at CERN. The analysis of multi-particle correlations was performed using the Hurst method. A detailed analysis of each individual event showed that in events of complete destruction of a projectile nucleus with a high multiplicity of secondary particles, long-distance multi-particle pseudo-rapidity correlations are observed. The distribution of average pseudo-rapidity in such events differs significantly from others, as it is much narrower, and its average value is noticeably shifted towards lower values <η>.

DrugDAGT: a dual-attention graph transformer with contrastive learning improves drug-drug interaction prediction

BackgroundDrug-drug interactions (DDIs) can result in unexpected pharmacological outcomes, including adverse drug events, which are crucial for drug discovery. Graph neural networks have substantially advanced our ability to model molecular representations; however, the precise identification of key local structures and the capture of long-distance structural correlations for better DDI prediction and interpretation remain significant challenges.ResultsHere, we present DrugDAGT, a dual-attention graph transformer framework with contrastive learning for predicting multiple DDI types. The dual-attention graph transformer incorporates attention mechanisms at both the bond and atomic levels, thereby enabling the integration of short and long-range dependencies within drug molecules to pinpoint key local structures essential for DDI discovery. Moreover, DrugDAGT further implements graph contrastive learning to maximize the similarity of representations across different views for better discrimination of molecular structures. Experiments in both warm-start and cold-start scenarios demonstrate that DrugDAGT outperforms state-of-the-art baseline models, achieving superior overall performance. Furthermore, visualization of the learned representations of drug pairs and the attention map provides interpretable insights instead of black-box results.ConclusionsDrugDAGT provides an effective tool for accurately predicting multiple DDI types by identifying key local chemical structures, offering valuable insights for prescribing medications, and guiding drug development. All data and code of our DrugDAGT can be found at https://github.com/codejiajia/DrugDAGT.

Research on Graph Multi-Attention Neural Network for Power System Transient Stability Assessment

Abstract Diagnosing the stability of power grids based on artificial intelligence technology is a research trend. The existing artificial intelligence stability analysis methods rely on a large number of fault case data, while the graph neural network method ignores the correlation characteristics of nodes themselves and the correlation of long-distance nodes. In order to solve the problems, the graph multi-attention neural network (GMANN) was proposed. The self-attention, which characterizes the correlation between different state quantities of a single node, is proposed, firstly. Then, long-distance association attention, which represents the correlation of distant nodes in the power grid, was proposed. The long-distance correlation attention and the adjacent correlation attention in graph attention network are fused to form a global attention that reflects the global importance of the power grid. The channel attention that characterizes the importance of different pooling methods of graph convolutional network is extracted and used to obtain the importance of different convolution operations. Finally, based on the multi-source attention fusion strategy, global attention and channel attention are embedded in the graph attention neural network to form a multi-level evaluation model to achieve power grid stability evaluation efficiently guided by multiple attentions. The proposed GMANN is verified based on simulated fault cases obtained from the 10-machine 39-node system in New England. The results show that the accuracy of the GMANN can reach 97.34%, which is better than other methods. And the missed judgment rate of important indicators is significantly better than other methods.

Dynamic category-sensitive hypergraph inferring and homo-heterogeneous neighbor feature learning for drug-related microbe prediction.

The microbes in human body play a crucial role in influencing the functions of drugs, as they can regulate the activities and toxicities of drugs. Most recent methods for predicting drug-microbe associations are based on graph learning. However, the relationships among multiple drugs and microbes are complex, diverse, and heterogeneous. Existing methods often fail to fully model the relationships. In addition, the attributes of drug-microbe pairs exhibit long-distance spatial correlations, which previous methods have not integrated effectively. We propose a new prediction method named DHDMP which is designed to encode the relationships among multiple drugs and microbes and integrate the attributes of various neighbor nodes along with the pairwise long-distance correlations. First, we construct a hypergraph with dynamic topology, where each hyperedge represents a specific relationship among multiple drug nodes and microbe nodes. Considering the heterogeneity of node attributes across different categories, we developed a node category-sensitive hypergraph convolution network to encode these diverse relationships. Second, we construct homogeneous graphs for drugs and microbes respectively, as well as drug-microbe heterogeneous graph, facilitating the integration of features from both homogeneous and heterogeneous neighbors of each target node. Third, we introduce a graph convolutional network with cross-graph feature propagation ability to transfer node features from homogeneous to heterogeneous graphs for enhanced neighbor feature representation learning. The propagation strategy aids in the deep fusion of features from both types of neighbors. Finally, we design spatial cross-attention to encode the attributes of drug-microbe pairs, revealing long-distance correlations among multiple pairwise attribute patches. The comprehensive comparison experiments showed our method outperformed state-of-the-art methods for drug-microbe association prediction. The ablation studies demonstrated the effectiveness of node category-sensitive hypergraph convolution network, graph convolutional network with cross-graph feature propagation, and spatial cross-attention. Case studies on three drugs further showed DHDMP's potential application in discovering the reliable candidate microbes for the interested drugs. Source codes and supplementary materials are available at https://github.com/pingxuan-hlju/DHDMP.

Unsupervised Social Bot Detection via Structural Information Theory

Research on social bot detection plays a crucial role in maintaining the order and reliability of information dissemination while increasing trust in social interactions. The current mainstream social bot detection models rely on black-box neural network technology, for example, Graph Neural Network, Transformer, and so on, which lacks interpretability. In this work, we present UnDBot, a novel unsupervised, interpretable, yet effective, and practical framework for detecting social bots. This framework is built upon structural information theory. We begin by designing three social relationship metrics that capture various aspects of social bot behaviors: posting type distribution , posting influence , and follow-to-follower ratio . Three new relationships are utilized to construct a new, unified, and weighted social multi-relational graph, aiming to model the relevance of social user behaviors and discover long-distance correlations between users. Second, we introduce a novel method for optimizing heterogeneous structural entropy. This method involves the personalized aggregation of edge information from the social multi-relational graph to generate a two-dimensional encoding tree. The heterogeneous structural entropy facilitates decoding of the substantial structure of the social bots network and enables hierarchical clustering of social bots. Third, a new community labeling method is presented to distinguish social bot communities by computing the user’s stationary distribution, measuring user contributions to network structure, and counting the intensity of user aggregation within the community. Compared with 10 representative social bot detection approaches, comprehensive experiments demonstrate the advantages of effectiveness and interpretability of UnDBot on 4 real social network datasets.

A mid-Cretaceous carbon isotope reference curve for the SE Neo-Tethys region (Zagros, W Iran)

Stable carbon isotope chemostratigraphy and the associated reference curves have proved to be valuable tools for stratigraphic calibration and long-distance correlations. Nonetheless, the reference curves covering the Cretaceous have mostly been produced from localities located within the W Neo-Tethys. In order to establish a reference curve for the SE Neo-Tethys margin, an Aptian–Albian hemipelagic carbonate-rich succession deposited along the Zagros Basin (SW Iran) was examined for planktic foraminifera biostratigraphy, carbon isotope stratigraphy, and sedimentological characteristics. Eight biozones including Globigerinelloides ferreolensis, Globigerinelloides algerianus, Hedbergella trocoidea, Paraticinella bejaouaensis, Muricohedbergella planispira, Ticinella (T.) primula, Biticinella breggiensis, and Pseudothalmanninella (P.)ticinensis as well as two subzones enclosing T. praeticinensis and P. subticinensis assign the strata to the early late Aptian–late Albian age. A carbon isotope record constrained by the planktic foraminifera zonation scheme shows good concordance with existing composite reference curves from the W Neo-Tethys region. Several carbon isotope excursions (CIEs) characterizing global marine events including Oceanic Anoxic Event (OAE) 1a, OAE1b, OAE1c and associated sub-events can be identified in the studied section. Deoxygenated bottom waters associated with the formation of OAE1a and OAE1b are indicated by distinct lithological signatures including enrichment in pyrite and glauconite, nodular and stratiform chert beds, as well as enhanced organic matter contents. We integrate the here-studied interval with a previously published upper Albian−Turonian succession from the same outcrop section, forming a composite, stratigraphically well-constrained section. This results in a high-resolution carbon isotope record for the SE Neo-Tethys margin, considered to represent an expedient reference Aptian–Turonian curve for this region. Correlation of this new curve with previously published records from the Middle East has helped to address ambiguities regarding the stratigraphic positions of the early/late Aptian and Aptian/Albian boundaries identified by previous studies.

Fault transformer: An automatic fault detection algorithm on seismic images using a transformer enhanced neural network

Seismic fault detection is a key step in seismic interpretation and reservoir characterization that often requires a large amount of human labor and interpretation time. Therefore, automatic seismic fault detection is critical for improving the efficiency of seismic data processing and interpretation. Existing artificial intelligence methods are mostly based on convolutional neural networks with a U-shaped encoder-decoder structure, known as U-net. However, the convolution is limited in modeling long-range correlative features. Instead, transformers, using self-attention mechanisms, avoid the local nature of the convolution, which has the potential to extract long-distance correlations. Transformers are proven to perform well in natural language processing, image classification, and segmentation tasks in precision and recall. Here, we develop a new deep neural network with transformers and a U-net-like structure: a fault transformer to perform the fault detection task. The new network outperforms the traditional U-net in the application with synthetic data sets.

Enhancement of alkaline water-oxidation activity for amorphous CoOx by the underlying remote-region atom seeding

Scaling up hydrogen production via large-current-density water electrolysis requires durable electrocatalysts of oxygen evolution reaction (OER) to reduce energy consumption. While amorphous metal oxides are promising OER catalysts, the surface-sites-electronic-orbital chance from the underlying remote-region atom seeding remains elusive. Herein, vanadium atom was controllable seated in the underlying remote region of the bulk materials as a typical coupling mode of amorphous-CoVOx-supported amorphous CoOx (CoVOx@CoOx), to study long-distance atoms correlation. XPS characterization for CoVOx@CoOx can't reveal the V signal due to the limited test depth of 5 nm, suggesting that V is existent in the internal region where it can be detected by EDS with 1 μm test depth. For surface-etched CoVOx@CoOx, V XPS signal appeared due to the exposure of CoVOx counterpart. The bionic flower structured, amorphous CoVOx@CoOx presents low overpotential of 338.5 mV at an extremely high current density of 1000 mA cm−2 in 1 M KOH and reserves 90.4% of the initial catalytical current after 24 h operation, significantly outperforming single amorphous CoOx. Density functional theory calculations evidence that the charge transfer from V atoms of CoVOx to surface active Co sites of CoOx part leads to the d-band center of Co-3d orbital shifted towards the Fermi level, which enhanced O adsorption effect thus promoting the OER activity. This research not only provides an ultrahigh-current-tolerance OER catalyst but also throws in-depth insight into the long-distance atoms charge interaction towards knowledge-guide catalyst construction for practical utilization.

VCAFusion: An infrared and visible image fusion network with visual perception and cross-scale attention

Infrared and visible image fusion methods aim to combine salient target instances and abundant texture details into fused images. However, due to the interference of harsh conditions, such as dense smoke, fog, and intense light, it is feasible for external interference information to be integrated into the fused image, which seriously affects the image quality. To this end, we propose an infrared and visible image fusion network with visual perception and cross-scale attention modules, termed VCAFusion, which integrates critical information from source images in harsh conditions more efficiently. Specifically, considering that the human eye can identify key information under adverse conditions, we design a visual perception module (VPM), guiding information integration from the perspective of human visual perception. In addition, we propose a cross-scale attention module (CSAM) based on shifted window cross-attention, which aims to capture the long-distance correlation between adjacent scale features, providing more accurate image information for image restoration. Experiments with a variety of datasets reveal that the VCAFusion can adaptively retain image information and further improve image generation ability in harsh conditions.

LCDFormer: Long-term correlations dual-graph transformer for traffic forecasting

Traffic forecasting has always been a critical component of intelligent transportation systems. Due to the complexity of traffic prediction models, most research just only consider short-term historical data in the temporal dimension. However, learning temporal patterns necessitates the involvement of long-term historical data. Additionally, many models are limited in capturing spatial features by only considering short-distance spatial information connected to the target node. To solve these problems, we propose a dual-graph transformer, namely Long-term Correlations Dual-graph transFormer (LCDFormer), designed to capture long-term correlations and long-distance spatial correlations. It is entirely based on attention mechanisms, and as far as we know, there is limited research adopting this approach. Our work addresses this gap in the literature. In particular, we have devised a time aggregation method capable of consolidating long-term historical time series, concurrently addressing the impact of long-term temporal correlations while minimizing the influence of redundant data. Subsequently, we have introduced a novel spatio-temporal attention module that compresses spatial information to generate short-term input sequences while modeling dynamic long-range spatial correlations. We conducted extensive experiments with LCDFormer on five real-world traffic datasets. The results indicate that LCDFormer, considering long-term spatio-temporal correlations, is better able to learn the spatio-temporal patterns of traffic data. Compared to the current state-of-the-art baseline, our model has demonstrated outstanding predictive performance with a maximum improvement of 5.02% in mean absolute error, 4.33% in root mean square error and 7.32% in mean absolute percentage error. The source codes are available at: https://github.com/NanakiC/LCDFormer.

QCD with an infrared fixed point: The pion sector

The possibility that gauge theories with chiral symmetry breaking below the conformal window exhibit an infrared fixed point is explored. With this assumption three aspects of pion physics are reproduced if the quark mass anomalous dimension at the infrared fixed point is γ*=1. First, by matching the long-distance scalar adjoint correlation function. Second, by perturbing the fixed point by a small quark mass, the mq-dependence of the pion mass is reproduced by renormalization group arguments. Third, consistency of the trace anomaly and the Feynman-Hellmann theorem, for small mq, imply the same result once more. This suggests the following picture for the conformal window; close to its upper boundary γ* is zero and grows as the number of fermions is reduced until its lower boundary γ*=1 is reached, where chiral symmetry breaking sets in. Below, the strongly coupled gauge theory with γ*=1 is infrared dual to the free theory of pions. A possible dilaton sector of the scenario will be addressed in a companion paper. Published by the American Physical Society 2024

STANet: Spatio-Temporal Adaptive Network and Clinical Prior Embedding Learning for 3D+T CMR Segmentation.

The segmentation of cardiac structure in magnetic resonance images (CMR) is paramount in diagnosing and managing cardiovascular illnesses, given its 3D+Time (3D+T) sequence. The existing deep learning methods are constrained in their ability to 3D+T CMR segmentation, due to: (1) Limited motion perception. The complexity of heart beating renders the motion perception in 3D+T CMR, including the long-range and cross-slice motions. The existing methods' local perception and slice-fixed perception directly limit the performance of 3D+T CMR perception. (2) Lack of labels. Due to the expensive labeling cost of the 3D+T CMR sequence, the labels of 3D+T CMR only contain the end-diastolic and end-systolic frames. The incomplete labeling scheme causes inefficient supervision. Hence, we propose a novel spatio-temporal adaptation network with clinical prior embedding learning (STANet) to ensure efficient spatio-temporal perception and optimization on 3D+T CMR segmentation. (1) A spatio-temporal adaptive convolution (STAC) treats the 3D+T CMR sequence as a whole for perception. The long-distance motion correlation is embedded into the structural perception by learnable weight regularization to balance long-range motion perception. The structural similarity is measured by cross-attention to adaptively correlate the cross-slice motion. (2) A clinical prior embedding learning strategy (CPE) is proposed to optimize the partially labeled 3D+T CMR segmentation dynamically by embedding clinical priors into optimization. STANet achieves outstanding performance with Dice of 0.917 and 0.94 on two public datasets (ACDC and STACOM), which indicates STANet has the potential to be incorporated into computer-aided diagnosis tools for clinical application.

Multi-scale topology and position feature learning and relationship-aware graph reasoning for prediction of drug-related microbes.

The human microbiome may impact the effectiveness of drugs by modulating their activities and toxicities. Predicting candidate microbes for drugs can facilitate the exploration of the therapeutic effects of drugs. Most recent methods concentrate on constructing of the prediction models based on graph reasoning. They fail to sufficiently exploit the topology and position information, the heterogeneity of multiple types of nodes and connections, and the long-distance correlations among nodes in microbe-drug heterogeneous graph. We propose a new microbe-drug association prediction model, NGMDA, to encode the position and topological features of microbe (drug) nodes, and fuse the different types of features from neighbors and the whole heterogeneous graph. First, we formulate the position and topology features of microbe (drug) nodes by t-step random walks, and the features reveal the topological neighborhoods at multiple scales and the position of each node. Second, as the features of nodes are high-dimensional and sparse, we designed an embedding enhancement strategy based on supervised fully connected autoencoders to form the embeddings with representative features and the more discriminative node distributions. Third, we propose an adaptive neighbor feature fusion module, which fuses features of neighbors by the constructed position- and topology-sensitive heterogeneous graph neural networks. A novel self-attention mechanism is developed to estimate the importance of the position and topology of each neighbor to a target node. Finally, a heterogeneous graph feature fusion module is constructed to learn the long-distance correlations among the nodes in the whole heterogeneous graph by a relationship-aware graph transformer. Relationship-aware graph transformer contains the strategy for encoding the connection relationship types among the nodes, which is helpful for integrating the diverse semantics of these connections. The extensive comparison experimental results demonstrate NGMDA's superior performance over five state-of-the-art prediction methods. The ablation experiment shows the contributions of the multi-scale topology and position feature learning, the embedding enhancement strategy, the neighbor feature fusion, and the heterogeneous graph feature fusion. Case studies over three drugs further indicate that NGMDA has ability in discovering the potential drug-related microbes. Source codes and Supplementary Material are available at https://github.com/pingxuan-hlju/NGMDA.

Learned Focused Plenoptic Image Compression With Microimage Preprocessing and Global Attention

Focused plenoptic cameras can record spatial and angular information of the light field (LF) simultaneously with higher spatial resolution relative to traditional plenoptic cameras, which facilitate various applications in computer vision. However, the existing plenoptic image compression methods present ineffectiveness to the captured images due to the complex micro-textures generated by the microlens relay imaging and long-distance correlations among the microimages. In this paper, a lossy end-to-end learning architecture is proposed to compress the focused plenoptic images efficiently. First, a data preprocessing scheme is designed according to the imaging principle to remove the sub-aperture image ineffective pixels in the recorded light field and align the microimages to the rectangular grid. Then, the global attention module with large receptive field is proposed to capture the global correlation among the feature maps using pixel-wise vector attention computed in the resampling process. Also, a new image dataset consisting of 1910 focused plenoptic images with content and depth diversity is built to benefit training and testing. Extensive experimental evaluations demonstrate the effectiveness of the proposed approach. It outperforms intra coding of HEVC and VVC by an average of 62.57% and 51.67% bitrate reduction on the 20 preprocessed focused plenoptic images, respectively. Also, it achieves 18.73% bitrate saving and generates perceptually pleasant reconstructions compared to the state-of-the-art end-to-end image compression methods, which benefits the applications of focused plenoptic cameras greatly. The dataset and code are publicly available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/VincentChandelier/GACN</uri> .

Hierarchical graph transformer with contrastive learning for gene regulatory network inference.

Gene regulatory networks (GRNs) are crucial for understanding gene regulation and cellular processes. Inferring GRNs helps uncover regulatory pathways, shedding light on the regulation and development of cellular processes. With the rise of high-throughput sequencing and advancements in computational technology, computational models have emerged as cost-effective alternatives to traditional experimental studies. Moreover, the surge in ChIPseq data for TF-DNA binding has catalyzed the development of graph neural network (GNN)-based methods, greatly advancing GRN inference capabilities. However, most existing GNN-based methods suffer from the inability to capture long-distance structural semantic correlations due to transitive interactions. In this paper, we introduce a novel GNN-based model named Hierarchical Graph Transformer with Contrastive Learning for GRN (HGTCGRN) inference. HGTCGRN excels at capturing structural semantics using a hierarchical graph Transformer, which introduces a series of gene family nodes representing gene functions as virtual nodes to interact with nodes in the GRNS. These semanticaware virtual-node embeddings are aggregated to produce node representations with varying emphasis. Additionally, we leverage gene ontology information to construct gene interaction networks for contrastive learning optimization of GRNs. Experimental results demonstrate that HGTCGRN achieves superior performance in GRN inference.

The critical role of accurate fossil identification: the case of the ostracods of the Codó Formation, NE Brazil

The Class Ostracoda comprises bivalve microcrustaceans with great diversity and abundance in the Lower Cretaceous paleolakes of Brazil and the world. With species generally well-preserved and of relatively short biochrons, it has provided high-resolution biostratigraphic frameworks that permit long-distance correlations. In this context, the Lower Cretaceous biozones of the NE Brazilian basins have allowed the establishment of correlations between coeval geological sections not only in this region but also with those of the Brazilian marginal basins, including the rocks of the pre-salt deposits, as well as with West African basins. In addition to biostratigraphy, these microfossils have proved indispensable in paleoenvironmental and paleogeographical analyses, including studies on the opening of the South Atlantic Ocean. However, some publications have emerged that, regrettably, exhibit errors in species identification and/or the accurate interpretation of biostratigraphic works. These mistakes have led to proposed modifications in both the biochron and geographic distribution of index taxa and taxonomic synonymizations that are incongruent with the current state of knowledge of the species. Consequently, these inaccuracies have given rise to erroneous biochronostratigraphic correlations and paleoenvironmental inferences. In this context, we meticulously examine the current understanding of the taxonomy of ostracods from the Codó Formation (NE Brazil), which is essential for the debate on the use of Lower Cretaceous ostracods in stratigraphy. Keywords: biostratigraphy, paleoecology, microfossils, Lower Cretaceous, Brazil.

Radial astrocyte synchronization modulates the visual system during behavioral-state transitions

Glial cells support the function of neurons. Recent evidence shows that astrocytes are also involved in brain computations. To explore whether and how their excitable nature affects brain computations and motor behaviors, we used two-photon Ca2+ imaging of zebrafish larvae expressing GCaMP in both neurons and radial astrocytes (RAs). We found that in the optic tectum, RAs synchronize their Ca2+ transients immediately after the end of an escape behavior. Using optogenetics, ablations, and a genetically encoded norepinephrine sensor, we observed that RA synchronous Ca2+ events are mediated by the locus coeruleus (LC)-norepinephrine circuit. RA synchronization did not induce direct excitation or inhibition of tectal neurons. Nevertheless, it modulated the direction selectivity and the long-distance functional correlations among neurons. This mechanism supports freezing behavior following a switch to an alerted state. These results show that LC-mediated neuro-glial interactions modulate the visual system during transitions between behavioral states.

Efficient multimodel method based on transformers and CoAtNet for Alzheimer's diagnosis

Convolutional neural networks (CNNs) have been widely used in medical imaging applications, including brain diseases such as Alzheimer's disease (AD) classification based on neuroimaging data. Researchers extract the potential brain regions related to AD disease using CNN from various imaging modalities due to its architectural inductive bias. The major limitation of the current CNN-based model is that it doesn't capture long-range relationships and long-distance correlation within the image features. Vision transformers (ViT) have proven an astounding performance in encoding long-range relationships with strong modeling capacity and global feature extraction due to the self attention mechanism. However, ViT doesn't model the spatial information or the local features within the image and is hard to train. Researchers have demonstrated that combining CNN and a transformer yields outstanding results. In this study, two new methods are proposed for Alzheimer's disease diagnosis. The first method combines the Swin transformer with an enhanced EfficientNet with multi-head attention and a Depthwise Over-Parameterized Convolutional Layer (DO-Conv). The second method consists of modifying the CoAtNet network with ECA-Net and fused inverted residuals blocks. We evaluated the effectiveness of our proposed methods based on the Open Access Series of Imaging Studies (OASIS) and the Alzheimer's Disease Neuroimaging Initiative (ADNI). Further, we evaluated the proposed methods using the Gradient-based Localization (Grad-CAM) method. The first method achieved 93.23% accuracy of classification on the OASIS dataset. The second method achieved 97.33% accuracy of classification on the OASIS dataset. We applied different multimodal image fusion methods (MRI and PET, MRI and CT) using our proposed method. The experimental results demonstrate that the fusion method based on PET and MRI outperforms the fusion method based on MRI and CT achieving 99.42% accuracy. Our methods outperform some traditional CNN models and the recent methods that are based on transformer for AD classification.

Point Cloud Completion Via Skeleton-Detail Transformer.

Point cloud shape completion plays a central role in diverse 3D vision and robotics applications. Early methods used to generate global shapes without local detail refinement. Current methods tend to leverage local features to preserve the observed geometric details. However, they usually adopt the convolutional architecture over the incomplete point cloud to extract local features to restore the diverse information of both latent shape skeleton and geometric details, where long-distance correlation among the skeleton and details is ignored. In this work, we present a coarse-to-fine completion framework, which makes full use of both neighboring and long-distance region cues for point cloud completion. Our network leverages a Skeleton-Detail Transformer, which contains cross-attention and self-attention layers, to fully explore the correlation from local patterns to global shape and utilize it to enhance the overall skeleton. Also, we propose a selective attention mechanism to save memory usage in the attention process without significantly affecting performance. We conduct extensive experiments on the ShapeNet dataset and real-scanned datasets. Qualitative and quantitative evaluations demonstrate that our proposed network outperforms current state-of-the-art methods.

The Local Unscented Transform Kalman Filter for the Weather Research and Forecasting Model

In this study, the local unscented transform Kalman filter (LUTKF) proposed in the previous study estimates the state of the Weather Research and Forecasting (WRF) model through local analysis. Real observations are assimilated to investigate the analysis performance of the WRF-LUTKF system. The WRF model as a regional numerical weather prediction (NWP) model is widely used to explain the atmospheric state for mesoscale meteorological fields, such as operational forecasting and atmospheric research applications. For the LUTKF based on the sigma-point Kalman filter (SPKF), the state of the nonlinear system is estimated by propagating ensemble members through the unscented transformation (UT) without making any linearization assumptions for nonlinear models. The main objective of this study is to examine the feasibility of mesoscale data assimilations for the LUTKF algorithm using the WRF model and real observations. Similar to the local ensemble transform Kalman filter (LETKF), by suppressing the impact of distant observations on model state variables through localization schemes, the LUTKF can eliminate spurious long-distance correlations in the background covariance, which are induced by the sampling error due to the finite ensemble size; therefore, the LUTKF used in the WRF-LUTKF system can efficiently execute the data assimilation with a small ensemble size. Data assimilation test results demonstrate that the LUTKF can provide reliable analysis performance in estimating the WRF model state with real observations. Experiments with various ensemble size show that the LETKF can provide better estimation results with a larger ensemble size, while the LUTKF can achieve accurate and reliable assimilation results even with a smaller ensemble size.