Search

Knowledge representation and reasoning (KRR) in complex systems (CSs) usually require facts from multiple experts having complementary backgrounds to fuse together. Consequently, such KRR methods should provide universal modeling languages close to human reasoning, with increased expressiveness and efficient capabilities to describe uncertainties. In this context, this paper introduces a new modeling formalism entitled Hybrid Logic-Algebraic Relational Modeling which is based on combining logic, probabilities, numerical information and network representations. The behavior, facts and workflows in a CS can be described using an environment of interconnected models enclosing sets of logical rules with attached probabilistic trust factors and links regarding logical attributes and numerical parameters. The logical and probabilistic inference applied to the modeling environment gives valuable knowledge to designers and decision-makers so that they can develop procedures or take actions in managing the CS. In this article, the proposed approach is completely formalized, from concept to definition and proofs and up to implementation, while its usage is illustrated within a complex economic, logistical, economical and technical scenario. • Describe and analyze workflows and reason decisions in complex systems. • Fuse together knowledge from actors with different fields of expertise. • Build intuition and identify global emergent particularities and constraints. • Fuse together logic rules, trust factors and numerical parameters. • Build networks of models using specific domain propagation mechanisms. [ABSTRACT FROM AUTHOR]

Silhouette is an effective modality in video-based person re-identification (ReID) since it contains features (e. g., stature and gait) complementary to the RGB modality. However, recent silhouette-assisted methods have not fully explored the spatial–temporal relations within each modality or considered the cross-modal complementarity in fusion. To address these two issues, we propose a Complete Relational Framework that includes two key components. The first component, Spatial-Temporal Relational Module (STRM), explores the spatiotemporal relations. STRM decomposes the video's spatiotemporal context into local/fine-grained and global/semantic aspects, modeling them sequentially to enhance the representation of each modality. The second component, Modality-Channel Relational Module (MCRM), explores the complementarity between RGB and silhouette videos. MCRM aligns two modalities semantically and multiplies them to capture complementary interrelations. With these two modules focusing on intra- and cross-modal relationships, our method achieves superior results across multiple benchmarks with minimal additional parameters and FLOPs. Code and models are available at https://github.com/baist/crf. • We consider silhouettes in video person re-identification to get robust features. • We explore both intra- and inter-modal relations for RGB and silhouette videos. • A novel decomposition strategy is proposed to improve efficiency. • Our method achieves state-of-the-art results in video person re-identification. [ABSTRACT FROM AUTHOR]

Analyzing the relationship between Internet attention and teaching research can provide a reference for extracting and identifying research hot spots in the discipline. Currently, many kinds of literature obtain social hot‐spot information based on a single platform, such as the Baidu Index, China National Knowledge Infrastructure (CNKI), and Web of Science. Still, they ignore the common information reflected between the Baidu Index platform and the professional academic platform and the lack of relationship demonstration between them. Therefore, based on the CNKI database and Baidu Index platform, the most critical data statistical analysis service platform in the Internet era, this article proposes a relationship modeling and prediction framework (RMPF). First, RMPF analyzes the relationship between public attention and teaching research from the perspective of "high school information technology" by word frequency analysis and Pearson. Second, RMPF constructs the relationship model between them using leave‐one‐out cross‐validation. Third, RMPF predicts the development of subject teaching and research using the autoregressive integrated moving average model and multiple regressions. The results show that the correlation coefficient between public attention and teaching research is more than 0.65 from the perspective of "high school information technology," showing a strong positive correlation. In recent years, the focus of high school information technology discipline tends to be integrating multimedia technology, cloud classroom, artificial intelligence, and other technologies and disciplines. In addition, discipline literature volume growth will slow in the next 2 years, with an average annual publication of 350 articles. The overall number of literature decreased by about 120 compared with 2020. Based on the proposed RMPF, this study clarifies the relationship between subject attention and teaching research through experimental demonstration and provides an implementation framework reference for researchers of relevant literature. [ABSTRACT FROM AUTHOR]

Recent studies of video action recognition can be classified into two categories: the appearance-based methods and the pose-based methods. The appearance-based methods generally cannot model temporal dynamics of large motion well by virtue of optical flow estimation, while the pose-based methods ignore the visual context information such as typical scenes and objects, which are also important cues for action understanding. In this paper, we tackle these problems by proposing a Pose-Appearance Relational Network (PARNet), which models the correlation between human pose and image appearance, and combines the benefits of these two modalities to improve the robustness towards unconstrained real-world videos. There are three network streams in our model, namely pose stream, appearance stream and relation stream. For the pose stream, a Temporal Multi-Pose RNN module is constructed to obtain the dynamic representations through temporal modeling of 2D poses. For the appearance stream, a Spatial Appearance CNN module is employed to extract the global appearance representation of the video sequence. For the relation stream, a Pose-Aware RNN module is built to connect pose and appearance streams by modeling action-sensitive visual context information. Through jointly optimizing the three modules, PARNet achieves superior performances compared with the state-of-the-arts on both the pose-complete datasets (KTH, Penn-Action, UCF11) and the challenging pose-incomplete datasets (UCF101, HMDB51, JHMDB), demonstrating its robustness towards complex environments and noisy skeletons. Its effectiveness on NTU-RGBD dataset is also validated even compared with 3D skeleton-based methods. Furthermore, an appearance-enhanced PARNet equipped with a RGB-based I3D stream is proposed, which outperforms the Kinetics pre-trained competitors on UCF101 and HMDB51. The better experimental results verify the potentials of our framework by integrating various modules. [ABSTRACT FROM AUTHOR]

Abstract Analyzing the relationship between Internet attention and teaching research can provide a reference for extracting and identifying research hot spots in the discipline. Currently, many kinds of literature obtain social hot‐spot information based on a single platform, such as the Baidu Index, China National Knowledge Infrastructure (CNKI), and Web of Science. Still, they ignore the common information reflected between the Baidu Index platform and the professional academic platform and the lack of relationship demonstration between them. Therefore, based on the CNKI database and Baidu Index platform, the most critical data statistical analysis service platform in the Internet era, this article proposes a relationship modeling and prediction framework (RMPF). First, RMPF analyzes the relationship between public attention and teaching research from the perspective of “high school information technology” by word frequency analysis and Pearson. Second, RMPF constructs the relationship model between them using leave‐one‐out cross‐validation. Third, RMPF predicts the development of subject teaching and research using the autoregressive integrated moving average model and multiple regressions. The results show that the correlation coefficient between public attention and teaching research is more than 0.65 from the perspective of “high school information technology,” showing a strong positive correlation. In recent years, the focus of high school information technology discipline tends to be integrating multimedia technology, cloud classroom, artificial intelligence, and other technologies and disciplines. In addition, discipline literature volume growth will slow in the next 2 years, with an average annual publication of 350 articles. The overall number of literature decreased by about 120 compared with 2020. Based on the proposed RMPF, this study clarifies the relationship between subject attention and teaching research through experimental demonstration and provides an implementation framework reference for researchers of relevant literature.

Graph Convolutional Network (GCN) which models the potential relationship between non-Euclidean spatial data has attracted researchers' attention in deep learning in recent years. It has been widely used in different computer vision tasks by modeling the latent space, topology, semantics, and other information in Euclidean spatial data and has achieved significant success. To better understand the work principles and future GCN applications in the computer vision field, this study reviewed the basic principles of GCN, summarized the difficulties and solutions using GCN in different visual tasks, and introduced in detail the methods for constructing graphs from the Euclidean spatial data in different visual tasks. At the same time, the review divided the application of GCN in basic visual tasks into image recognition, object detection, semantic segmentation, instance segmentation and object tracking. The role and performance of GCN in basic visual tasks were summarized and compared in detail for different tasks. This review emphasizes that the application of GCN in computer vision faces three challenges: computational complexity, the paradigm of constructing graphs from the Euclidean spatial data, and the interpretability of the model. Finally, this review proposes two future trends of GCN in the vision field, namely model lightweight and fusing GCN with other models to improve the performance of the visual model and meet the higher requirements of vision tasks. [ABSTRACT FROM AUTHOR]

Estimating human poses from videos is critical in human–computer interaction. Joints cooperate rather than move independently during human movement. There are both spatial and temporal correlations between joints. Despite the positive results of previous approaches, most of them focus on modeling the spatial correlation between joints while only straightforwardly integrating features along the temporal dimension, which ignores the temporal correlation between joints. In this work, we propose a plug-and-play kinematics modeling module (KMM) to explicitly model temporal correlations between joints across different frames by calculating their temporal similarity. In this way, KMM can capture motion cues of the current joint relative to all joints in different time. Besides, we formulate video-based human pose estimation as a Markov Decision Process and design a novel kinematics modeling network (KIMNet) to simulate the Markov Chain, allowing KIMNet to locate joints recursively. Our approach achieves state-of-the-art results on two challenging benchmarks. In particular, KIMNet shows robustness to the occlusion. Code will be released at https://github.com/YHDang/KIMNet. • We propose a plug-and-play kinematics modeling module to explicitly model temporal correlations between joints across frames. • We formulate video-based human pose estimation as a Markov Decision Process and present a KIMNet to simulate the Markov Chain. • The proposed KIMNet locates the occluded joint by integrating joints' information from other frames. • The proposed KIMNet achieves state-of-the-state results and shows superior performance on the occluded scenes. [ABSTRACT FROM AUTHOR]

Road traffic is an important factor restricting economic development. With the rapid growth of the number of motor vehicles, traffic accidents, traffic congestion and traffic pollution have become the persistent diseases of contemporary society, and become more and more serious. The existing traffic accident analysis is difficult to comprehensively and objectively reflect the process of traffic accidents. In order to improve the effectiveness of traffic accident prevention and measures, it is particularly important to carry out road traffic accident analysis. The paper proposes a new method for modeling the relationship between vehicle speed dispersion and traffic accidents in mixed traffic flow. Simulation results verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Recent work has identified correlations between early mastery of fractions and later math achievement, especially in algebra. However, causal connections between aspects of reasoning with fractions and improved algebra performance have yet to be established. The current study investigated whether relational reasoning with fractions facilitates subsequent algebraic reasoning using both pre-algebra students and adult college students. Participants were first given either a relational reasoning fractions task or a fraction algebra procedures control task. Then, all participants solved word problems and constructed algebraic equations in either multiplication or division format. The word problems and the equation construction tasks involved simple multiplicative comparison statements such as “There are 4 times as many students as teachers in a classroom.” Performance on the algebraic equation construction task was enhanced for participants who had previously completed the relational fractions task compared with those who completed the fraction algebra procedures task. This finding suggests that relational reasoning with fractions can establish a relational set that promotes students’ tendency to model relations using algebraic expressions. [ABSTRACT FROM AUTHOR]

Abstract: We describe a simple computing technique for solving independence and domination problems on rectangular chessboards. It rests upon relational modeling and uses the BDD-based specific purpose computer algebra system RelView for the evaluation of the relation-algebraic expressions that specify the problems’ solutions and the visualization of the computed results. The technique described in the paper is very flexible and especially appropriate for experimentation. It can easily be applied to other chessboard problems. [Copyright &y& Elsevier]

Abstract: This paper presents a formal verification methodology of high-level data-flow synthesis process. Typically, given a data-flow description, the high-level data-flow synthesis tools perform high-level restructuring and data-flow transformations to produce an optimized architectural implementation. How to verify the correctness of the synthesis results is a key issue within the high-level data-flow synthesis process. Our approach adopts the relational modeling techniques from Kleene algebra with tests (KAT) and the symbolic computation methods from polynomial algebra to solve this verification problem. In this paper, we show how to extract target properties from the data-flow description by using KAT and how to construct the polynomial representation for the architectural implementation. Further, we demonstrate how a symbolic computation-based decision procedure is integrated into our framework to perform verification in a generalized bounded model checking (BMC) style. The experimental results on some public benchmark and practical designs demonstrate the efficiency of our approach and its applicability to large data-flow designs. [Copyright &y& Elsevier]