Search

Introduction: Accurate recognition of martial arts leg poses is essential for applications in sports analytics, rehabilitation, and human-computer interaction. Traditional pose recognition models, relying on sequential or convolutional approaches, often struggle to capture the complex spatial-temporal dependencies inherent in martial arts movements. These methods lack the ability to effectively model the nuanced dynamics of joint interactions and temporal progression, leading to limited generalization in recognizing complex actions. Methods: To address these challenges, we propose PoseGCN, a Graph Convolutional Network (GCN)-based model that integrates spatial, temporal, and contextual features through a novel framework. PoseGCN leverages spatial-temporal graph encoding to capture joint motion dynamics, an action-specific attention mechanism to assign importance to relevant joints depending on the action context, and a self-supervised pretext task to enhance temporal robustness and continuity. Experimental results on four benchmark datasets—Kinetics-700, Human3.6M, NTU RGB+D, and UTD-MHAD—demonstrate that PoseGCN outperforms existing models, achieving state-of-the-art accuracy and F1 scores. Results and discussion: These findings highlight the model's capacity to generalize across diverse datasets and capture fine-grained pose details, showcasing its potential in advancing complex pose recognition tasks. The proposed framework offers a robust solution for precise action recognition and paves the way for future developments in multi-modal pose analysis. [ABSTRACT FROM AUTHOR]

Deep learning has gained superior accuracy on Euclidean structure data in neural networks. As a result, non-Euclidean structure data, such as graph data, has more sophisticated structural information, which can be applied in neural networks as well to address more complex and practical problems. However, actual graph data obeys a power-law distribution, so the adjacent matrix of a graph is random and sparse. Graph processing accelerator (GPA) is designed to handle the problems above. However, graph computing only processes 1-dimensional data. In graph neural networks (GNNs), graph data is multi-dimensional. Consequently, GNNs include the execution processes of both traditional graph processing and neural network, which have irregular memory access and regular computation, respectively. To obtain more information in graph data and require better model generalization ability, the layers of GNN are deeper, so the overhead of memory access and computation is considerable. At present, GNN accelerators are designed to deal with this issue. In this paper, we conduct a systematic survey regarding the design and implementation of GNN accelerators. Specifically, we review the challenges faced by GNN accelerators, and existing related works in detail to process them. Finally, we evaluate previous works and propose future directions in this booming field. [ABSTRACT FROM AUTHOR]

Aspect-Based Sentiment Analysis (ABSA) targets sentiments on specific aspects in reviews, offering more granularity than overall sentiment analysis. Challenges in ABSA include handling implicit sentiments, varying expressions, linguistic nuances, and ensuring robust predictions across domains. Addressing these is crucial for extracting meaningful insights from customer reviews and enhancing products or services. Aiming at these concerns, this paper proposes an Enhanced Knowledge Infused Graph-Gated BERT (EKIG-GBERT) model for ABSA in customer-related program reviews. This innovative approach integrates a Dynamic Sentiment-specific Knowledge Graph (DSSKG) and Knowledge graph-enhanced BERT model with Gated Domain Graph Convolutional Network (KG-BERT-GDGCN) to capture intricate sentiment-aspect relationships. The methodology begins with data pre-processing, including tokenization and noise reduction, followed by domain-specific knowledge infusion via DSSKG. The approach leverages KG-BERT for advanced aspect extraction, enhancing the model's capacity to capture subtle emotional nuances in textual data. Aspect extraction is performed at multiple levels like term, category, implicit, entity, and attribute that leverages the KG-BERT model for comprehensive sentiment representation. Additionally, a structured graph seamlessly integrates affective information from DSSKG and KG-BERT, forming an affective adjacency matrix that encapsulates nuanced emotional connections among words in a sentence. The integrated sentiment prediction framework fuses features from DSSKG and KG-BERT using the GDGCN model. Processing through densely connected layers, dropout, and batch normalization ensures effective regularization, resulting in a robust model that leverages information from multiple sources for improved sentiment analysis. Experimental evaluations using four SemEval datasets (i.e., Rest14 task 4, Lap14 task 4, Res15 task 12, Res16 task 5) demonstrate that the EKIG-GBERT model significantly outperforms existing ABSA methods. The EKIG-GBERT model achieved an accuracy of 97.5% on the Rest14 task 4, 98.5% on Lap14 task 4, 94% on Res15 task 12, and 92% on Res16 task 5. Additionally, the confusion matrix analysis further confirmed its superior performance in distinguishing between various sentiment aspects. These results underscore the model's robustness and reliability in the ABSA prediction tasks. [ABSTRACT FROM AUTHOR]

Real-time and accurate traffic forecasting aids in traffic planning and design and helps to alleviate congestion. Addressing the negative impacts of partial data loss in traffic forecasting, and the challenge of simultaneously capturing short-term fluctuations and long-term trends, this paper presents a traffic forecasting model, D-MGDCN-CLSTM, based on Multi-Graph Gated Dilated Convolution and Conv-LSTM. The model uses the DTWN algorithm to fill in missing data. To better capture the dual characteristics of short-term fluctuations and long-term trends in traffic, the model employs the DWT for multi-scale decomposition to obtain approximation and detail coefficients. The Conv-LSTM processes the approximation coefficients to capture the long-term characteristics of the time series, while the multiple layers of the MGDCN process the detail coefficients to capture short-term fluctuations. The outputs of the two branches are then merged to produce the forecast results. The model is tested against 10 algorithms using the PeMSD7(M) and PeMSD7(L) datasets, improving MAE, RMSE, and ACC by an average of 1.38% and 13.89%, 1% and 1.24%, and 5.92% and 1%, respectively. Ablation experiments, parameter impact analysis, and visual analysis all demonstrate the superiority of our decompositional approach in handling the dual characteristics of traffic data. [ABSTRACT FROM AUTHOR]