Search

In the study of human action recognition, two-stream networks have made excellent progress recently. However, there remain challenges in distinguishing similar human actions in videos. This paper proposes a novel local-aware spatio-temporal attention network with multi-stage feature fusion based on compact bilinear pooling for human action recognition. To elaborate, taking two-stream networks as our essential backbones, the spatial network first employs multiple spatial transformer networks in a parallel manner to locate the discriminative regions related to human actions. Then, we perform feature fusion between the local and global features to enhance the human action representation. Furthermore, the output of the spatial network and the temporal information are fused at a particular layer to learn the pixel-wise correspondences. After that, we bring together three outputs to generate the global descriptors of human actions. To verify the efficacy of the proposed approach, comparison experiments are conducted with the traditional hand-engineered IDT algorithms, the classical machine learning methods (i.e., SVM) and the state-of-the-art deep learning methods (i.e., spatio-temporal multiplier networks). According to the results, our approach is reported to obtain the best performance among existing works, with the accuracy of 95.3% and 72.9% on UCF101 and HMDB51, respectively. The experimental results thus demonstrate the superiority and significance of the proposed architecture in solving the task of human action recognition.

Public space is characterised by visible spaces and featured by observable activities, forming the atmosphere that humans can perceive, interpret, and experience. This article conceptualises urban space as a set of viewsheds connecting functions visually which can be represented as functional visibility graphs – the graphs with mutual-visibility edges between space and function nodes. It begins with an introduction of three basic measures: visual function size, entropy, and mean angular depth step, then proposes advanced measures: namely visual function connectivity, function visibility, and visible function closeness, showing spatial narratives along paths, visual centrality in place, and functional regions for continuous areas, respectively. This framework is enabled by social media check-in data that records people’s engagements across function nodes tagged by them. An application in a real example in Tianjin City is shown. The novelty of using social media check-ins as a delegation of actual function usage is demonstrated for modelling urban movement with improved precision. By tracing the shifting performance of the functional visibility graphs for the same spatial layout, this study outlines how such analysis can be conducted for uncovering short-term transformation of the visual landscape, contributing to the fine-scale, high-resolution implementations of land use policies from a real human-focused perspective, with the socially sensed data.

In this paper, the province construction and demolition waste (CDW) generation in China from 2005 to 2018 is calculated by using the area estimation method. Together with the use of the gravity model and the social network analysis method (SNA), the characteristics and effects of the spatial correlation structure network of China's province construction and demolition waste generation are studied. It has been found that from 2005 to 2018, the construction and demolition waste in China had seen a significant rise, and the spatial difference in province generation is significant. During that time, China's inter-provincial construction and demolition waste generation demonstrated significant characteristics of a spatial correlation network with a complex network hierarchy and showed a trend of year-by-year increase in its overall network density and number of correlation. In this spatial connection network, nine provinces, including Beijing, Tianjin, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Hubei and Guangdong, played a core role as "bridges", Beijing, Shanghai, Jiangsu, Guangdong, Shandong, Fujian and Zhejiang provinces are "beneficiaries", Chongqing, Tianjin, Hubei, Anhui, Sichuan, Henan, Jiangxi, Hunan and Hebei are "brokers" while other provinces are "the damaged". Indexes such as network density, network hierarchy and network efficiency all have a significant impact on the national construction and demolition waste generation intensity. The reduction of network density, the improvement of network hierarchy and network efficiency can significantly reduce the difference in construction and demolition waste generation intensity. The structure of spatial correlation network has significant influence on the difference of construction and demolition waste generation intensity.

Mixed networks of hydrogen bonds in a water–monoethanolamine (MEA) liquid system formed from solvents with a spatial network of hydrogen bonds are described topologically. This allows the comparison of networks of different systems and selection of the best one (particularly for cryobiological problems). The coordinates of atoms in the Н2О–MEA system are obtained via molecular dynamics (MD). The mixed network is described by means of graph theory and Delaunay simplices at 300 K throughout the range of concentration of monoethanolamine in water, based on MD results. Earlier works and published data on structure and interaction in the system are used to explain the observations. The amine group of a MEA molecule plays a key role in the formation of the MEA⋅H2O stable complex in the mixed network of H-bonds of the system in the entire concentration range. The graph theory results are compared to ones obtained with Delaunay simplices. The results are compared to similar studies of the mixed network of a H2O–EG system.

Research on urban agglomerations from the perspective of network spatial structure is important to promote their sustainable development. Based on online and traditional data, this paper first improves three aspects of the traditional spatial gravity model—city quality, the gravitation coefficient, and city distance—considering urban center functional intensity and population mobility tendencies. The resulting improved directional gravity model is applied to analyze the structure of the city network for two urban agglomerations in China, the Beijing-Tianjin-Hebei urban agglomeration (BTHUA) and the Yangtze River Delta urban agglomeration (YRDUA). The results of the study are as follows: (1) the existing urban connections have obvious hierarchies and imbalances, with the YRDUA urban hierarchical connections being of larger scale. (2) Cities are closely connected, but city networks are unbalanced, though the YRDUA has more balanced urban development. (3) Each node city has a clear radiation range limit, and spatial distance remains an important constraint on urban connections. The backbone network of the BTHUA has a triangular shape and trends toward a “sparse north and dense south,” while the YRDUA is characterized by multiple axes and an overall distribution that trends toward a “dense north and sparse south.” (4) Cities with poor comprehensive strength are more likely to be captured, forming an attract and be attracted relationship. (5) The BTHUA and the YRDUA each form three communities.

We present a framework for solving simultaneously the problems of facility location and path optimization in static and dynamic spatial networks. In the static setting, the objective is to determine facility locations and transportation paths from each node to the destination via the network of facilities such that the total cost of commodity transportation is minimized. This is an NP-hard problem. We propose a novel stage-wise viewpoint of the paths which is instrumental in designing the decision variable space in our framework. We use the maximum entropy principle to solve the resulting optimization problem. In the dynamic setting, nodes and destinations are dynamic. We design an appropriate control Lyapunov function to determine the time evolution of facilities and paths such that the transportation cost at each time instant is minimized. Our framework enables quantifying attributes of the facilities and transportation links in terms of the decision variables. Consequently, it becomes possible to incorporate application specific constraints on individual facilities, links, and network topology. We demonstrate the efficacy of our proposed framework through extensive simulations.

The naïve algorithm for generating nearest-neighbour models determines the distance between every pair of nodes, resulting in quadratic running time. Such time complexity is common among spatial problems and impedes the generation of larger spatial models. In this article, an improved algorithm for the Mocnik model, an example of nearest-neighbour models, is introduced. Instead of solving k nearest-neighbour problems for each node (k dynamic in the sense that it varies among the nodes), the improved algorithm presented exploits the notion of locality through introducing a corresponding spatial index, resulting in a linear average-case time complexity. This makes possible to generate very large prototypical spatial networks, which can serve as testbeds to evaluate and improve spatial algorithms, in particular, with respect to the optimization of algorithms towards big geospatial data.

Ultra-dense network (UDN) is a promising solution to meet the huge data requirements of future 5G wireless communication systems. With many irregularly deployed small cell base stations (SBSs), it is very difficult to accurately measure network energy efficiency (EE) in UDN. Considering the necessity of saving energy in modern-day networks, in this paper, I present the EE evaluation of UDN based on the random spatial network model. In this considered system, the SBSs and the macro cell users (MUEs) are modeled as a homogeneous Poisson point process (PPP). Unlike the existing literature, I derive the simple closed-form expressions for the successful transmission probabilities of MUEs and small cells by Laplace transformation. Then, based on the obtained successful transmission probabilities, I derive the closed form expression for EE of the small cells in both uplink and downlink under some practical constraints. Further, to obtain the optimal SBS density in UDN, I propose a joint uplink and downlink power allocation (JUDPA) algorithm. The JUDPA algorithm obtains the maximum EE in small cells by iteratively optimizing the uplink and downlink powers. Our theoretical analysis and simulated results demonstrate the accuracy of the proposed model with better power allocation and less complexity than conventional solutions.

We study boundary conditions for the diffusion operator defined on a star-shaped geometric graph consisting of three edges with a common vertex. We show that if the edge lengths are pairwise distinct, then there do not exist degenerate boundary conditions for the diffusion operator. If the edge lengths coincide and the potentials are symmetric, then the characteristic determinant of a boundary value problem for the diffusion operator cannot be a constant other than zero, and the set of boundary value problems for which the characteristic determinant is identically zero is infinite (a continuum). We show that, for the diffusion operator on the star-shaped graph, the set of boundary value problems whose spectrum fills the entire plane consists of eighteen classes, each of which contains eight to nine arbitrary constants. Recall that for the diffusion operator defined on an interval this set consists of two problems.

Recent geometric deep learning works define convolution operations in local regions and have enjoyed remarkable success on non-Euclidean data, including graph and point clouds. However, the high-level geometric correlations between the input and its neighboring coordinates or features are not fully exploited, resulting in suboptimal segmentation performance. In this article, we propose a novel graph convolution architecture, which we term as Taylor Gaussian mixture model (GMM) network (TGNet), to efficiently learn expressive and compositional local geometric features from point clouds. The TGNet is composed of basic geometric units, TGConv, that conduct local convolution on irregular point sets and are parametrized by a family of filters. Specifically, these filters are defined as the products of the local point features and the neighboring geometric features extracted from local coordinates. These geometric features are expressed by Gaussian weighted Taylor kernels. Then, a parametric pooling layer aggregates TGConv features to generate new feature vectors for each point. TGNet employs TGConv on multiscale neighborhoods to extract coarse-to-fine semantic deep features while improving its scale invariance. Additionally, a conditional random field (CRF) is adopted within the output layer to further improve the segmentation results. Using three point cloud data sets, qualitative and quantitative experimental results demonstrate that the proposed method achieves 62.2% average accuracy on ScanNet, 57.8% and 68.17% mean intersection over union (mIoU) on Stanford Large-Scale 3D Indoor Spaces (S3DIS) and Paris-Lille-3D data sets, respectively.

In various wireless networking settings, node locations determine a network’s topology, allowing the network to be modelled by a geometric graph drawn in the plane. Without any additional information, local geometric routing algorithms can guarantee delivery to the target node only in restricted classes of geometric graphs, such as triangulations. In order to guarantee delivery on more general classes of geometric graphs (e.g., convex subdivisions or planar subdivisions), previous local geometric routing algorithms required [Formula: see text] state bits to be stored and passed with the message. We present the first local geometric routing algorithm using only one state bit to guarantee delivery on convex subdivisions and, when the algorithm has knowledge of the incoming port (the preceding node on the route), the first stateless local geometric routing algorithm that guarantees delivery on edge-augmented monotone subdivisions (including all convex subdivisions). We also show that [Formula: see text] state bits are necessary in planar subdivisions in which faces may have three or more reflex vertices.

Multisource aggregation (MSA) is an important function in the remote sensing data processing. In this letter, we propose and study a novel MSA search and scheduling problem to improve the performance of remote sensing data cluster. Given a static spatial network, a set of query points $Q$ , and a set of data locations $O$ (candidates for cluster centers), the MSA function retrieves the data location with the minimum aggregation distance (the sum of the distances to all query points). We believe that such function plays an important role in remote sensing data cluster and classification. The MSA problem faces two challenges: 1) how to prune the search space effectively and retrieve the results of MSA in real time and 2) how to schedule multiple query sources during search processing. To overcome these challenges, we make the following contributions. First, upper and lower bounds on aggregation distance are defined to prune the search space effectively. Second, each query source is given a priority label, and a best-first scheduling strategy is developed to further enhance the query performance. Finally, we conduct extensive experiments on real and synthetic data sets to verify the high performance of the developed algorithms.

FOSS4G 2021 Buenos Aires, Buenos Aires, Argentine, 27-/09/2021 - 02/10/2021; Arabesque is an application for the exploration and geovisualisation of origin-destination flows (or spatial networks), developed within the framework of the Univ. Gustave Eiffel (ex. IFSTTAR)-funded research project geographic flow visualisation (gflowiz) geoflowiz, in collaboration with the CNRS. It allows both the exploration and the filtering of OD data and their representation, with a strong emphasis on geographic information layering and features' semiology. The key-objective is to propose an easy way to produce a modern cartography (a geovisualisation) of thematic flows (e.g. bilateral flow volume), at several geographic scales, even from your own datasets. The objective of this article is to position Arabesque in the range of geoweb applications for producing flow maps, by comparing its functionalities with those of similar web applications- Magrit, Kepler.gl, flowmap.blue -pointing out their respective advantages and limitations. The analysis of its functionalities is compared on the same flow dataset -MOBSCO, i.e. a dataset describing the school mobility of French pupils and students on a given year- for a practical and empirical "validation" of its contributions. We demonstrate that the configurations and appearances of these tools' visual output depend largely on the culture of their developers, and on the use and audiences for which they have been developed.

Under the tide of global economic integration with aggravated environmental risks caused by intensive human activities, the spatial network correlation of environmental risks has become intensified. The close exchange of activities among the countries under the Belt and Road Initiative (BRI) will especially feel this contagion of environmental risks. Given this situation, this article analyzes the environmental risk contagion relations among the BRI countries and the characteristics of their network structure by using social network analysis (SNA). A block model is used to analyze the spatial clustering characteristics of the environmental risk contagion. Specifically, the driving factors of environmental risk contagion are analyzed through the quadratic assignment procedure (QAP) of SNA. The results of this article provide some references for the BRI to reduce the environmental risks and jointly control the environmental risk contagion, so as to assist in the promotion of a green silk road.

Inspired by the great success of the deep neural networks in various fields of computer vision, studies for image deblurring have begun to become more active in recent days. However, most previous approaches often fail to accurately remove the blur artifacts, e.g., ghosting effects at the object boundaries and degradation of local details, in restored results. In this paper, we propose a deep spectral-spatial network (DSSN) for resolving the problem of single image deblurring. Specifically, the proposed method is able to efficiently recover scene characteristics in a global manner by minimizing differences of the frequency magnitude between the blurred input and corresponding sharp image via the spectral restorer, and the spatial restorer fine-tunes local details of the intermediate result, which is estimated by the spectral one, based on the intensity similarity. This cascaded scheme of deblurring processes is fairly desirable for clearly restoring edge-like structures as well as the textural information in a coarse-to-fine manner. Experimental results on benchmark datasets demonstrate that the proposed DSSN outperforms state-of-the-art methods. The code and model are publicly available at: https://github.com/SeokjaeLIM/DSSN_release .

This paper provides evidence about the relevant role of geography on spatial local networks in trade credit between industrial reduced size companies and distin- guishing the spatial behavior accor...

Finding a maximal independent set ( MIS ) in a graph is one of the fundamental problems in distributed computing. Researchers in this community are trying to close the time complexity gap of computing a MIS in a graph, way back from Luby’s [STOC’85] O ( log n ) time (randomized) algorithm and Linial’s [SICOMP’92] Ω ( log ∗ n ) lower bound result ( n is the number of nodes of the graph). Since then, an extensive research has been done on this problem, however, most of the results are randomized and we are lack of efficient deterministic solution. In fact, no polylogarithmic (in n ) time deterministic algorithm is known so far in general graphs — an open question raised by Linial in 1993 [Comb. Probab. Comput.’93]. In this paper, we study the MIS problem in geometric graphs, a class of graphs which has both theoretical and practical importance. We present O ( 1 ) -time deterministic algorithms (hence, optimal) for computing MIS in unit interval graphs, unit square graphs and unit disk graphs. The same idea applies to develop optimal MIS algorithm for higher dimensional unit balls and unit hypercubes. We further extend the MIS algorithms to compute approximate maximum independent set ( MaxIS ) in the above geometric graph classes. The theoretical results are corroborated through extensive experiments to show the effectiveness and efficiency of our algorithms. Our algorithms are fully decentralized, scalable, and robust. In general, nodes exchange only O ( log n ) bits message through an edge per communication. Moreover, being local, our algorithms are also suitable in failure model, self-stabilizing and appropriate for dynamic environment.

We study conditions for the solvability of boundary value problems for differential equations of arbitrary order on a geometric graph. The boundary conditions are given by func-tionals that are linear combinations of the one-sided limits of the solution and its derivatives calculated at all graph vertices. The dimensions of the linear spaces of solutions of homogeneous mutually adjoint boundary value problems are proved to be the same. Conditions for the solvability and unique solvability of a degenerate boundary value problem are established. A generalized Green’s function is constructed, its uniqueness is proved, and its properties are described. A theorem on the uniform convergence of the sequence of solutions of the degenerate boundary value problem under the condition of uniform convergence of its right-hand sides is proved.

A key technical issue for human intention prediction from observed human actions is how to discover and utilize the spatio-temporal patterns behind those actions. Inspired by the well-known two-stream architecture for action recognition, this paper proposes an approach for human intention prediction based on a two-stream architecture using RGB images and optical flow. Firstly, the action-start frame of each trial of a human action is determined by calculating the L2 distance of the positions of the human joints between frames of the skeleton data. Secondly, a spatial network and a temporal network are trained separately to predict human intentions using RGB images and optical flow, respectively. Both early concatenation fusion methods in the spatial network and sampling methods in the temporal network are optimized based on experiments. Finally, average fusion is used to fuse the prediction results from the spatial network and the temporal network. To verify the effectiveness of the proposed approach, a new dataset of human intentions behind human actions is introduced. This dataset contains RGB images, RGB-D images, and skeleton data of human actions of pitching a ball. Experiments show that the proposed approach can predict human intentions behind human actions with a prediction accuracy of 74% on the proposed dataset. The proposed approach is further evaluated on the Intention from Motion (IfM) dataset, a dataset of human intentions behind human actions of grasping a bottle. The proposed approach achieves a prediction accuracy of 77% on the IfM dataset. The proposed approach is effective in predicting human intentions behind human actions in different applications.

In real world, cooperation among unrelated agents remains to represent one open challenge in many disciplines. In some cases, subjects can even change their social dilemmas to ensure their own benefit. Inspired by this fact, here we propose a multigame (composed of prisoner's dilemma game and snowdrift game) and its coevolution mechanism in networked populations: if a player succeeds to study the strategy of its opponent, it also learns the game type of its opponent. Based on numerous computation simulations, it is unveiled that compared with the setup of multigame, coevolution mechanism can effectively resolve the problem of collective cooperation. While for this observation, it is attributed to the fact that vast majority of players changed their games to snowdrift game (i.e. lower social dilemma), where cooperation-defection bistable state guarantees cooperation clusters. We hope our finding can inspire more studies for resolving the social dilemmas.

A geometric graph is called self-approaching, if for each (ordered) pair of vertices of the graph, there exists a self-approaching (directed) path between them. A path from a vertex u to a vertex v of the graph is called self-approaching, if for any point (not just vertex) x on the path, the distance between x and any point that starts at u and moves toward x is decreasing. A path is called an increasing-chord, if it is self-approaching from both sides, source to destination and destination to source. A geometric graph is called a self-approaching t-spanner, for t > 1 , if for each pair ( u , v ) of vertices of the graph, there exists a self-approaching path from u to v of length at most t times the Euclidean distance between u and v. In 2015, Dehkordi et al. (2015) [16] left the problem of the existence of an increasing-chord planar geometric graph for a set of points that lie along the sides of an acute triangle, as an open problem, but they did not mention anything about the case of an obtuse triangle. In this paper, we show that for each set of points on the boundary of an obtuse triangle, there exists an increasing-chord planar geometric graph. Moreover, we show that for each set of points on the boundary of an acute triangle, there exists an increasing-chord planar geometric graph using at most three Steiner points. We also introduce weakly self-approaching geometric graphs (spanners) as a variation of self-approaching geometric graphs (spanners). A weakly self-approaching path is defined similarly to a self-approaching path, except that when a point p on the path moves from the source to the destination, the Euclidean distance between p and the destination vertex (but not all the points on the path) decreases. A weakly self-approaching geometric graph (spanner) is defined by replacing “self-approaching” by “weakly self-approaching” in the above definition. In this paper, we show that for every point set S ⊆ R d , there is a weakly self-approaching planar geometric graph spanning S. Furthermore, this study shows how to test in polynomial time, whether a given geometric graph is weakly self-approaching. Notably, the corresponding decision problem in the context of self-approaching geometric graphs in R d with d ≥ 3 is NP-hard. We also propose some algorithms for constructing a weakly self-approaching geometric t-spanner for a given point set and a real constant t > 1 . Also, we show that for every point set S on the boundary of any triangle, there exists an increasing-chord t-spanner spanning S and a weakly self-approaching t-spanner spanning S, both with O ( k n ) edges, in which k depends only on t.

Complex network theory is a multidisciplinary research direction of complexity science which has experienced a rapid surge of interest over the last two decades. Its applications in land-based urban traffic network studies have been fruitful, but have suffered from the lack of a systematic cognitive and integration framework. This paper reviews complex network theory related knowledge and discusses its applications in urban traffic network studies in several directions. This includes network representation methods, topological and geographical related studies, network communities mining, network robustness and vulnerability, big-data-based research, network optimization, co-evolution research and multilayer network theory related studies. Finally, new research directions are pointed out. With these efforts, this physics-based concept will be more easily and widely accepted by urban traffic network planners, designers, and other related scholars.

We study the boundary conditions of the Sturm-Liouville problem posed on a star-shaped geometric graph consisting of three edges with a common vertex. We show that the Sturm-Liouville problem has no degenerate boundary conditions in the case of pairwise distinct edge lengths. However, if the edge lengths coincide and all potentials are the same, then the characteristic determinant of the Sturm-Liouville problem cannot be a nonzero constant and the set of Sturm-Liouville problems whose characteristic determinant is identically zero and whose spectrum accordingly coincides with the entire plane is infinite (a continuum). It is shown that, for one special case of the boundary conditions, this set consists of eighteen classes, each having from two to four arbitrary constants, rather than of two problems as in the case of the Sturm-Liouville problem on an interval.

The inverse Sturm-Liouville problem with nonseparated boundary conditions on a star-shaped geometric graph consisting of three edges with a common vertex is studied. It is shown that the Sturm-Liouville problem with general boundary conditions cannot be reconstructed uniquely from four spectra. A class of nonseparated boundary conditions is obtained for which two uniqueness theorems for the solution of the inverse Sturm-Liouville problem are proved. In the first theorem, the data used to reconstruct the Sturm-Liouville problem are the spectrum of the boundary value problem itself and the spectra of three auxiliary problems with separated boundary conditions. In the second theorem, instead of the spectrum of the problem itself, one only deals with five of its eigenvalues. It is shown that the Sturm-Liouville problem with these nonseparated boundary conditions can be reconstructed uniquely if three spectra of auxiliary problems and five eigenvalues of the problem itself are used as the reconstruction data. Examples of unique reconstruction of potentials and boundary conditions of the Sturm-Liouville problem posed on the graph under study are given.

Temporal graphs are the modeling of pairwise and historical interaction in recordings of a dataset. A temporal matching formalizes the planning of pair working sessions of a required duration. We depict algorithms finding temporal matchings maximizing the total workload, by an exact algorithm and an approximation. The exact algorithm is a dynamic programming solving the general case in \(O^*((\gamma +1)^n)\) time, where n is the number of vertices, \(\gamma \) represents the desired duration of each pair working session, and \(O^*\) only focuses on exponential factors. When the input data is embedded in an Euclidean space, called geometric data, our approximation is based on a new notion of temporal velocity. We revise a known notion of static density [van Leeuwen, 2009] and result in a polynomial time approximation scheme for temporal geometric graphs of bounded density. We confront our implementations to known opensource implementation (Our source code is available at https://github.com/Talessseed/Temporal-matching-of-historical-and-geometric-graphs).

Predicting how changes to the urban environment layout will affect the spatial distribution of pedestrian flows is important for environmental, social and economic sustainability. We present longitudinal evaluation of a model of the effect of urban environmental layout change in a city centre (Cardiff 2007–2010), on pedestrian flows. Our model can be classed as regression based direct demand using Multiple Hybrid Spatial Design Network Analysis (MH-sDNA) assignment, which bridges the gap between direct demand models, facility-based activity estimation and spatial network analysis (which can also be conceived as a pedestrian route assignment based direct demand model). Multiple theoretical flows are computed based on retail floor area: everywhere to shops, shop to shop, railway stations to shops and parking to shops. Route assignment, in contrast to the usual approach of shortest path only, is based on a hybrid of shortest path and least directional change (most direct) with a degree of randomization. The calibration process determines a suitable balance of theoretical flows to best match observed pedestrian flows, using generalized cross-validation to prevent overfit. Validation shows that the model successfully predicts the effect of layout change on flows of up to approx. 8000 pedestrians per hour based on counts spanning a 1 km2city centre, calibrated on 2007 data and validated to 2010 and 2011. This is the first time, to our knowledge, that a pedestrian flow model with assignment has been evaluated for its ability to forecast the effect of urban layout changes over time.

This paper presents a new framework to classify floor plan elements and represent them in a vector format. Unlike existing approaches using image-based learning frameworks as the first step to segment the image pixels, we first convert the input floor plan image into vector data and utilize a graph neural network. Our framework consists of three steps. (1) image pre-processing and vectorization of the floor plan image, (2) region adjacency graph conversion, and (3) the graph neural network on converted floor plan graphs. Our approach is able to capture different types of indoor elements including basic elements, such as walls, doors, and symbols, as well as spatial elements, such as rooms and corridors. In addition, the proposed method can also detect element shapes. Experimental results show that our framework can classify indoor elements with an F1 score of 95%, with scale and rotation invariance. Furthermore, we propose a new graph neural network model that takes the distance between nodes into account, which is a valuable feature of spatial network data.

The spatial distribution of population affects disease transmission, especially when shelter in place orders restrict mobility for a large fraction of the population. The spatial network structure of settlements therefore imposes a fundamental constraint on the spatial distribution of the population through which a communicable disease can spread. In this analysis we use the spatial network structure of lighted development as a proxy for the distribution of ambient population to compare the spatiotemporal evolution of COVID-19 confirmed cases in the USA and China. The Visible Infrared Imaging Radiometer Suite (VIIRS) Day/Night Band sensor on the NASA/NOAA Suomi satellite has been imaging night light at ~ 700 m resolution globally since 2012. Comparisons with sub-kilometer resolution census observations in different countries across different levels of development indicate that night light luminance scales with population density over ~ 3 orders of magnitude. However, VIIRS’ constant ~ 700 m resolution can provide a more detailed representation of population distribution in peri-urban and rural areas where aggregated census blocks lack comparable spatial detail. By varying the low luminance threshold of VIIRS-derived night light, we depict spatial networks of lighted development of varying degrees of connectivity within which populations are distributed. The resulting size distributions of spatial network components (connected clusters of nodes) vary with degree of connectivity, but maintain consistent scaling over a wide range (5 × to 10 × in area & number) of network sizes. At continental scales, spatial network rank-size distributions obtained from VIIRS night light brightness are well-described by power laws with exponents near −2 (slopes near −1) for a wide range of low luminance thresholds. The largest components (104to 105km2) represent spatially contiguous agglomerations of urban, suburban and periurban development, while the smallest components represent isolated rural settlements. Projecting county and city-level numbers of confirmed cases of COVID-19 for the USA and China (respectively) onto the corresponding spatial networks of lighted development allows the spatiotemporal evolution of the epidemic (infection and detection) to be quantified as propagation within networks of varying connectivity. Results for China show rapid nucleation and diffusion in January 2020 followed by rapid decreases in new cases in February. While most of the largest cities in China showed new confirmed cases approaching zero before the end of February, most of these cities also showed distinct second waves of cases in March or April. Whereas new cases in Wuhan did not approach zero until mid-March, as of December 2020 it has not yet experienced a second wave of cases. In contrast, the results for the USA show a wide range of trajectories, with an abrupt transition from slow increases in confirmed cases in a small number of network components in January and February, to rapid geographic dispersion to a larger number of components shortly before mobility reductions occurred in March. Results indicate that while most of the upper tail of the network had been exposed by the end of March, the lower tail of the component size distribution has only shown steep increases since mid-June.

This paper focuses on modeling a disorder ensemble of quantum dots (QDs) as a special kind of Random Geometric Graphs (RGG) with weighted links. We compute any link weight as the overlap integral (or electron probability amplitude) between the QDs (=nodes) involved. This naturally leads to a weighted adjacency matrix, a Laplacian matrix, and a time evolution operator that have meaning in Quantum Mechanics. The model prohibits the existence of long-range links (shortcuts) between distant nodes because the electron cannot tunnel between two QDs that are too far away in the array. The spatial network generated by the proposed model captures inner properties of the QD system, which cannot be deduced from the simple interactions of their isolated components. It predicts the system quantum state, its time evolution, and the emergence of quantum transport when the network becomes connected.

International audience; We are given the adjacency matrix of a geometric graph and the task of recovering the latent positions. We study one of the most popular approaches which consists in using the graph distances and derive error bounds under various assumptions on the link function. In the simplest case where the link function is an indicator function, the bound is (nearly) optimal as it (nearly) matches an information lower bound.

In video action recognition based on deep learning, the design of the neural network is focused on how to acquire effective spatial information and motion information quickly. This paper proposes a kind of deep network that can obtain both spatial information and motion information in video classification. It is called MDFs (the multidimensional motion features of deep feature map net). This method can be used to obtain spatial information and motion information in videos only by importing image frame data into a neural network. MDFs originate from the definition of 3D convolution. Multiple 3D convolution kernels with different information focuses are used to act on depth feature maps so as to obtain effective motion information at both spatial and temporal. On the other hand, we split the 3D convolution at space dimension and time dimension, and the spatial network feature map has reduced the dimensions of the original frame image data, which realizes the mitigation of computing resources of the multichannel grouped 3D convolutional network. In order to realize the region weight differentiation of spatial features, a spatial feature weighted pooling layer based on the spatial-temporal motion information guide is introduced to realize the attention to high recognition information. By means of multilevel LSTM, we realize the fusion between global semantic information acquisition and depth features at different levels so that the fully connected layers with rich classification information can provide frame attention mechanism for the spatial information layer. MDFs need only to act on RGB images. Through experiments on three universal experimental datasets of action recognition, UCF10, UCF11, and HMDB51, it is concluded that the MDF network can achieve an accuracy comparable to two streams (RGB and optical flow) that requires the import of both frame data and optical flow data in video classification tasks.

Given a collection of multi-attribute trajectories, an event definition, and a spatial network, the Significant Lagrangian Linear Hotspot Discovery (SLLHD) problem finds the paths where records in the trajectories tend to be events in the Lagrangian perspective. The SLLHD problem is of significant societal importance because of its applications in transportation planning, vehicle design, and environmental protection. Its main challenges include the potentially large number of candidate hotspots caused by the tremendous volume of trajectories as well as the non-monotonicity of the statistic measuring event concentration. The related work on the linear hotspot discovery problem is designed in the Eulerian perspective and focuses on point datasets, which ignores the dependence of event occurrence on trajectories and the paths where trajectories are. To solve this problem, we introduce an algorithm in the Lagrangian perspective, as well as five refinements that improve its computational scalability. Two case studies on real-world datasets and experiments on synthetic data show that the proposed approach finds hotspots which are not detectable by existing techniques. Cost analysis and experimental results on synthetic data show that the proposed approach yields substantial computational savings.

We consider 1-local online routing on a special class of geometric graphs called Delaunay triangulations (DTs). A geometric graph G = (V, E) of a point set consists of a set of points in the plane and edges between them, where each edge weighs as the Euclidean distance between it’s end-points. DTs are one of the useful classes of these graphs because of some good properties which can help during the navigation process, therefore over the years DTs have been widely proposed as network topologies for several times.In this paper, we present a new memoryless online routing (MOR) algorithm for DTs which is simple, elegant, and easy to implement, while having an acceptable performance.The set of MOR algorithms are suitable for cases where we want to find a path using only local information, our proposed algorithm is memoryless or 1-local, in k-local routing, we find a path between a source vertex s to a destination vertex t while our knowledge at each step is limited to the locations of s and t, the location of current vertex and it’s k-neighborhood vertices.We also evaluate and compare the perforamnce of our prpopsed algorithm with existing MOR algorithms. Our experimental results implied that our proposed algorithm has an acceptable performance in both Euclidean and link metrics and it outperforms all of the existing MOR algorithms in Euclidean metric, and some of them in the link metric as well. Finally, we pose two open problems to solve in the future.

In this study, we propose novel centrality measures considering multiple perspectives of nodes or node groups based on the facility location problem on a spatial network. The conventional centrality exclusively quantifies the global properties of each node in a network such as closeness and betweenness, and extracts nodes with high scores as important nodes. In the context of facility placement on a network, it is desirable to place facilities at nodes with high accessibility from residents, that is, nodes with a high score in closeness centrality. It is natural to think that such a property of a node changes when the situation changes. For example, in a situation where there are no existing facilities, it is expected that the demand of residents will be satisfied by opening a new facility at the node with the highest accessibility, however, in a situation where there exist some facilities, it is necessary to open a new facility some distance from the existing facilities. Furthermore, it is natural to consider that the concept of closeness differs depending on the relationship with existing facilities, cooperative relationships and competitive relationships. Therefore, we extend a concept of centrality so as to considers the situation where one or more nodes have already been selected belonging to one of some groups. In this study, we propose two measures based on closeness centrality and betweenness centrality as behavior models of people on a spatial network. From our experimental evaluations using actual urban street network data, we confirm that the proposed method, which introduces the viewpoints of each group, shows that there is a difference in the important nodes of each group viewpoint, and that the new store location can be predicted more accurately.

In order to effectively evaluate personnel performance, a distributed data mining algorithm for spatial networks based on BP neural wireless network is proposed. In the cloud computing environment, an excavator is used to construct multiple input multiple output spatial network data, analyze the data structure, and perform redundant data compression of massive data through time-frequency feature extraction. Combined with adaptive matching filtering method, the characteristics of the data are matched. The spatial frequency feature extraction method is used to locate the features of the multiple-input multiple-output spatial network data. In order to improve the accuracy of data mining, the BP neural network is used to classify and identify the extracted data features to achieve the optimization of data mining. A wireless sensor network is a wireless network composed of a large number of stationary or moving sensors in a self-organizing and multi-hop manner. It cooperatively senses, collects, processes, and transmits the information of the perceived objects in the geographical area covered by the network, and finally puts these The information is sent to the owner of the network. This algorithm improves the accuracy of personnel performance evaluation, and simultaneously establishes a hierarchical analysis and quantitative evaluation model for the performance of government managers, and adjusts the results of hierarchical statistical analysis on government administrators as needed. The performance evaluation and optimization of government administrators were introduced. The empirical analysis results show that the method has higher accuracy for government managers' performance evaluation, higher efficiency of big data processing and better integration.

Effective video representation is a key ingredient in action recognition, but how to learn effective spatial features remains a fundamental and challenging task. The existing CNN-based methods apply low-resolution feature maps to get the high-level semantic labels. However, the slenderer spatial information for action representation has lost. In this paper, we propose a novel stacked spatial network (SSN), which integrates multi-layer feature maps in an end-to-end manner. Spatial features extraction network based on encoder-decoder is firstly used to obtain multi-level and multi-resolution spatial features under the supervision of high-level sematic labels. The multi-level features are aggregated through a stacked spatial fusion layer, which intrinsically refines the traditional convolutional neural network. Then, refined spatial network (RSN) is proposed to aggregate spatial network and SSN. Particularly, the learned representation of RSN comprises two components for representing semantic label information and local slenderer spatial information. Extensive experimental results on UCF-101 and HMDB-51 datasets demonstrate the effectiveness of the proposed RSN.

In the present paper, we investigate the initial boundary value problem describing oscillation process on a geometric graph with hysteresis type boundary conditions. The analogue of the d'A...

This paper examines the spatially stratified association between various urban leisure entertainment activities and street configurations via a spatial design network analysis and attempts to provide insightful implications for the sustainable development of a city by using a typical inland city of China, Wuhan, as an example. The street network is characterized by performing a spatial network analysis of the closeness, betweenness, severance and efficiency parameters. The kernel density estimation (KDE) method is employed to measure the proximities of every place to various types of leisure entertainment facilities. The two sets of densities are then analyzed to determine whether spatially stratified correlations exist and whether various leisure entertainment facilities are more spatially correlated with specific street features. The results indicate that a good street network design can be measured not only by closeness or betweenness but also by severance and efficiency. Four street network metrics are spatially correlated with the locations of leisure entertainment activities. Various types of leisure entertainment activities have distinct locational preferences for street networks (e.g., chess and card rooms favor closeness). The conclusions of this paper can help urban planners and policy-makers spatially optimize street networks and leisure venues to develop a sustainable city.

A regional coordinated development strategy is an important measure that is often used to implement sustainable development in China. However, many obstacles greatly limit the realization of regional ecological coordinated sustainable development. In this paper, ecological efficiency is utilized as an important indicator of sustainable development, and the network analysis method is used to explore the spatial correlation relationship of regional ecological coordinated sustainable development. This paper calculates the ecological efficiency of each region using the Window slacks-based measure (Window-SBM) model, formulates the spatial network of regional ecological efficiency spillover through the vector auto-regressive (VAR) Granger causality model, and analyzes the spatial spillover relationship and influencing factors of regional ecological efficiency by using the social network analysis method. It is found that the spillover network of ecological efficiency in each region presents a typical core-edge structure. In addition, there is an obvious hierarchical structure among blocks with different directions and functions. Industrial structure, economic development, and geographical proximity have a positive impact on the spatial spillover of regional ecological efficiency, while environmental regulation has a negative impact. Finally, relevant policy suggestions are put forward.

Many real-world spatial systems can be conceptualized as networks. In these conceptualizations, nodes and links represent system components and their interactions, respectively. Traditional network analysis applies graph theory measures to static network datasets. However, recent interest lies in the representation and analysis of evolving networks. Existing network automata approaches simulate evolving network structures, but do not consider the representation of evolving networks embedded in geographic space nor integrating actual geospatial data. Therefore, the objective of this study is to integrate network automata with geographic information systems (GIS) to develop a novel modelling framework, Geographic Network Automata (GNA), for representing and analyzing complex dynamic spatial systems as evolving geospatial networks. The GNA framework is implemented and presented for two case studies including a spatial network representation of (1) Conway&rsquo, s Game of Life model and (2) Schelling&rsquo, s model of segregation. The simulated evolving spatial network structures are measured using graph theory. Obtained results demonstrate that the integration of concepts from geographic information science, complex systems, and network theory offers new means to represent and analyze complex spatial systems. The presented GNA modelling framework is both general and flexible, useful for modelling a variety of real geospatial phenomena and characterizing and exploring network structure, dynamics, and evolution of real spatial systems. The proposed GNA modelling framework fits within the larger framework of geographic automata systems (GAS) alongside cellular automata and agent-based modelling.

Spatial network analysis is a collection of methods for measuring accessibility potentials as well as for analyzing flows over transport networks. Though it has been part of the practice of geographic information systems for a long time, designing network analytical workflows still requires a considerable amount of expertise. In principle, artificial intelligence methods for workflow synthesis could be used to automate this task. This would improve the (re)usability of analytic resources. However, though underlying graph algorithms are well understood, we still lack a conceptual model that captures the required methodological know-how. The reason is that in practice this know-how goes beyond graph theory to a significant extent. In this article we suggest interpreting spatial networks in terms of quantified relations between spatial objects, where both the objects themselves and their relations can be quantified in an extensive or an intensive manner. Using this model, it becomes possible to effectively organize data sources and network functions towards common analytical goals for answering questions. We tested our model on 12 analytical tasks, and evaluated automatically synthesized workflows with network experts. Results show that standard data models are insufficient for answering questions, and that our model adds information crucial for understanding spatial network functionality.

We study the problem of computing Euclidean minimum weight matching which investigates the minimum weight perfect matching in a complete geometric graph on a set of 2n points in the plane. We propose a simple randomized approximation algorithm based on the geometrical aspect of the problem with O(n) expected time. The proposed algorithm computes a matching within at most 3 factors of the optimal solution. We also do some experimental tests to evaluate the performance of the proposed algorithm which indicate the efficiency of the proposed algorithm in finding the approximate matching in the practice.

City governments all over the world face challenges understanding mobility patterns within dense urban environments at high spatial and temporal resolution. While such measures are important to provide insights into the functional patterns of a city, novel quantitative methods, derived from ubiquitous mobile connectivity, are needed to provide policy-makers with better insights to improve urban management and planning decisions. In this paper, we develop a model that uses large-scale WiFi probe request data to model urban mobility trajectories in dense urban environments. We collect probe request data from a public Wifi network with 54 access points in the Lower Manhattan section of New York City over one week, accounting for more than 30 million observations and over 800,000 unique devices. First, we aggregate unique entries per access point and per hour, demonstrating the potential to use WiFi data to approximate local population counts by type of user. We then use a spatial network analysis to identify edge frequencies and directions of journeys between the network nodes, and apply the results to the road and pedestrian sidewalk network to identify usage intensity levels and trajectories for individual street segments. We demonstrate the significant potential in the use of WiFi probe request data for understanding mobility patterns in cities, while highlighting non-trivial issues in data privacy raised by the growing availability of public WiFi networks.

The lack of access to electricity in developing countries necessitates spatial electricity planning for guiding sustainable electrification projects that evaluate the costs of centralized systems vis-a-vis decentralized systems. Heuristic approaches have been widely used in such electrification problems to find feasible, cost effective solutions; however, most of the time global optimality of these solutions is not guaranteed. Our paper through its modeling approach provides a new methodology to find the least cost solution to this electrification problem. We model the spatial network planning problem as Prize Collecting Steiner Tree problem, which would be a base for a decision support tool for rural electrification. This new method is systematically assessed using both randomly generated data and real data from rural regions across Sub-Saharan Africa. Comparative results for the proposed approach and a widely used heuristic method are presented based on computational experiments.

The relationship between cross-regional cultural landscapes does not currently receive enough attention in cultural landscape conservation. Cultural landscapes in China are faced with the crisis of fragmentation and islanding, which makes it necessary to strengthen the idea of integrated conservation. The Great Wall cultural belt in Beijing is a cross-regional linear cultural landscape in the northwest mountainous area of Beijing. With the Great Wall as a link connecting the surrounding natural, social, and cultural resources to form a banded structure, it has the typical and practical significance of integrated conservation. Based on the theory of the cultural landscape security pattern and its ‘node – setting – connection’ model, this research studies the spatial structure and shaping mechanism of the Great Wall cultural belt in Beijing from the perspective of tourist flow, using Flickr geo-tagged photos and Wikiloc tracks. The results identify the core nodes and intermediate nodes of the Great Wall cultural belt in Beijing and show that it has a multi-core, multi-group hierarchical nesting structure differing from the geographic space, which has groups following the evolution orbit of ‘linear string – circumnuclear star – complete network.’ To explain the differences between the cultural association network (CAN) and the geospatial network (GSN), we find that the cultural associations between nodes rely on geophysical channels, including heritage corridors and traffic passages, as physical carriers, but the nodes’ attraction and their setting elements, containing natural geography environments, settlements, administrative divisions, etc., change the channel resistance, so CAN and GSN present different characteristics. From the perspective of systematic and spatial network thoughts, this study has made an attempt to adopt new analysis and research methods to achieve the integrated conservation and inheritance of linear cultural landscapes in mountainous areas.

The concept of an Ω-weakly regular problem is introduced. On the basis of the Zhuravlev operator approach combined with the neural network paradigm, it is shown that, for each such problem, a correct algorithm and a six-level spatial neural network reproducing the computations executed by this algorithm can be constructed. Moreover, the set of Ω-weakly regular problems includes the set of Ω-regular problems. It turns out that a three-level spatial network (μ-block) is a forward propagation network whose inner loop under estimation of the class membership for each test object consists of a single iteration. As a result, the amount of computations required for the six-level network is reduced noticeably.

To address the effects of increasing homeless populations, planners must understand the size and distribution of their homeless populations, as well as how information and resources are diffused throughout homeless communities. Currently, there is limited publicly available information on the homeless population, e.g. the estimates of the homeless, gathered annually by the US Housing & Urban Development point in time survey. While it is theorized in the literature that the networks of homeless individuals provide access to important information for planners in areas such as health (e.g. needle exchanges) or access (e.g. information diffusion about the location of new shelters), it is almost never measured, and if measured, only at a very small scale. This research addresses the question of how planners can leverage publicly available data on the homeless to better understand their own homeless networks (e.g. relations among the homeless themselves) in a cost-effective and reliable way. To this end, we provide a method for simulating realistic networks of a social relation among the homeless population and perform a diffusion analysis over the resultant homeless-to-homeless networks, and also over a simulated homeless youth Facebook network. We validate the former through novel use of historical data, while the latter is based on recent work that demonstrated that the homeless youth have similar size Facebook networks and usage. We see much stronger spatial hopping and quicker diffusion over the youth network, i.e. we expect information to pass among the youth network much faster than the homeless-to-homeless network. This finding implies that non-government organizations and public health efforts that seek to provide information, goods or services to the homeless should start with the homeless youth, given the potential for faster diffusion when homeless youth are the initial transmitters. Overall, these methods and analysis provide a unique opportunity for visualizing, characterizing and inferring information for large-scale and hard to measure social networks.

This study suggests a new spatial network model, called the space-connector model, for improving the representation of relations between spaces. Previous models have proved incapable of expressing ...

In this paper, we propose a topology preserving graph matching (TPGM) method for partial face recognition. Most existing face recognition methods extract features from holistic facial images. However, faces in real-world unconstrained environments may be occluded by objects or other faces, which cannot provide the whole face images for description. Keypoint-based partial face recognition methods such as multi-keypoint descriptor with Gabor ternary pattern and robust point set matching match the local keypoints for partial face recognition. However, they simply measure the nodewise similarity without higher order geometric graph information, which are susceptible to noises. To address this, our TPGM method estimates a non-rigid transformation encoding the second-order geometric structure of the graph, so that more accurate and robust correspondence can be computed with the topological information. In order to exploit higher order topological information, we propose a topology preserving structural matching method to construct a higher order structure for each face and estimate the transformation. Experimental results on four widely used face data sets demonstrate that our method outperforms most existing state-of-the-art face recognition methods.

The super-resolution problem for hyperspectral images is currently one of the most challenging topics in remote sensing. Increasingly effective methods have been presented to solve this ill-posed p...