Enqing Chen’s research while affiliated with Zhengzhou University and other places

Graph convolutional networks (GCNs) are an effective skeleton-based human action recognition (HAR) technique. GCNs enable the specification of CNNs to a non-Euclidean frame that is more flexible. The previous GCN-based models still have a lot of issues: (I) The graph structure is the same for all model layers and input data.

Quantum Human Action Recognition (HAR) is an interesting research area in human-computer interaction used to monitor the activities of elderly and disabled individuals affected by physical and mental health. In the recent era, skeleton-based HAR has received much attention because skeleton data has shown that it can handle changes in striking, body size, camera views, and complex backgrounds. One key characteristic of ST-GCN is automatically learning spatial and temporal patterns from skeleton sequences. It has some limitations, as this method only works for short-range correlation due to its limited receptive field. Consequently, understanding human action requires long-range interconnection. To address this issue, we developed a quantum spatial-temporal relative transformer ST-RTR model. The ST-RTR includes joint and relay nodes, which allow efficient communication and data transmission within the network. These nodes help to break the inherent spatial and temporal skeleton topologies, which enables the model to understand long-range human action better. Furthermore, we combine quantum ST-RTR with a fusion model for further performance improvements. To assess the performance of the quantum ST-RTR method, we conducted experiments on three skeleton-based HAR benchmarks: NTU RGB+D 60, NTU RGB+D 120, and UAV-Human. It boosted CS and CV by 2.11 % and 1.45% on NTU RGB+D 60, 1.25% and 1.05% on NTU RGB+D 120. On UAV-Human datasets, accuracy improved by 2.54%. The experimental outcomes explain that the proposed ST-RTR model significantly improves action recognition associated with the standard ST-GCN method.

5-G/6G technology improves skeleton-based human action recognition (HAR) by delivering ultra-low latency and high data throughput for real-time and accurate security analysis of human actions. Despite its growing popularity, current HAR methods frequently fail to capture the skeleton sequence’s complexities. This study proposes a novel multimodal method that synergizes the Spatial-Temporal Attention LSTM (STA-LSTM) Network with the Convolutional Neural Network (CNN) to extract nuanced features from the skeleton sequence. The STA-LSTM network dives deep into inter-and intra-frame relations, while the CNN model uncovers geometric correlations within the human skeleton. Significantly, by integrating the Choquet fuzzy integral, we achieve a harmonized fusion of classifiers for each feature vector. Adopting Kullback Leibler and Jensen-Shannon divergences further ensures the complementary nature of these feature vectors. STA-LSTM Network and CNN in the proposed multimodal method significantly advance human action recognition. Impressive accuracy was demonstrated by our approach after evaluating benchmark skeletal datasets such as NTU-60, NTU-120, HDM05, and UT-DMHAD. Specifically, it achieved C-subject 90.75%, 84.50%, and C-setting 96.7% and 86.70% on NTU-60 and NTU-120, respectively. Furthermore, HDM05 and UT-DMHAD datasets recorded accuracies of 93.5% and 97.43%, indicating that our model outperforms current techniques and has excellent potential for sentiment analysis platforms that combine textual and visual signals.

Developing complex hyperspectral image (HSI) sensors that capture high-resolution spatial information and voluminous (hundreds) spectral bands of the earth’s surface has made HSI pixel-wise classification a reality. The 3D-CNN has become the preferred HSI pixel-wise classification approach because of its ability to extract discriminative spectral and spatial information while maintaining data integrity. However, HSI datasets are characterized by high nonlinearity, voluminous spectral features, and limited training sample data. Therefore, developing deep HSI classification methods that purely utilize 3D-CNNs in their network structure often results in computationally expensive models prone to overfitting when the model depth increases. In this regard, this paper proposes an integrated deep multi-scale 3D/2D convolutional network block (MiCB) for simultaneous low-level spectral and high-level spatial feature extraction, which can optimally train on limited sample data. The strength of the proposed MiCB model solely lies in the innovative arrangement of convolution layers, giving the network the ability (i) to simultaneously convolve the low-level spectral with high-level spatial features; (ii) to use multiscale kernels to extract abundant contextual information; (iii) to apply residual connections to solve the degradation problem when the model depth increases beyond the threshold; and (iv) to utilize depthwise separable convolutions in its network structure to address the computational cost of the proposed MiCB model. We evaluate the efficacy of our proposed MiCB model using three publicly accessible HSI benchmarking datasets: Salinas Scene (SA), Indian Pines (IP), and the University of Pavia (UP). When trained on small amounts of training sample data, MiCB is better at classifying than the state-of-the-art methods used for comparison. For instance, the MiCB achieves a high overall classification accuracy of 97.35%, 98.29%, and 99.20% when trained on 5% IP, 1% UP, and 1% SA data, respectively.

Human action recognition using skeletons has become increasingly appealing to a growing number of researchers in recent years. It is particularly challenging to recognize actions when they are captured from different angles because there are so many variations in their representations. This paper proposes an automatic strategy for determining virtual observation viewpoints that are based on learning and data driven to solve the problem of view variation throughout an act. Our VA-CNN and VA-RNN networks, which use convolutional and recurrent neural networks with long short-term memory, offer an alternative to the conventional method of reorienting skeletons according to a human-defined earlier benchmark. Using the unique view adaption module, each network first identifies the best observation perspectives and then transforms the skeletons for end-to-end detection with the main classification network based on those viewpoints. The suggested view adaptive models can provide significantly more consistent virtual viewpoints using the skeletons of different perspectives. By removing views, the models allow networks to learn action-specific properties more efficiently. Furthermore, we developed a two-stream scheme (referred to as VA-fusion) that integrates the performance of two networks to obtain an improved prediction. Random rotation of skeletal sequences is used to avoid overfitting during training and improve the reliability of view adaption models. An extensive experiment demonstrates that our proposed view adaptive networks outperform existing solutions on five challenging benchmarks.

Human Action Recognition (HAR) using skeletons has become increasingly appealing to a growing number of researchers in recent years. It is particularly challenging to recognize actions when they are captured from different angles because there are so many variations in their representations. This paper proposes an automatic strategy for determining virtual observation viewpoints that are based on learning and data-driven to solve the problem of view variation throughout an act. Our VA-CNN and VA-RNN networks, which use convolutional and recurrent neural networks with the Long Short-term Memory (LSTM), offer an alternative to the conventional method of reorienting skeletons according to a human-defined earlier benchmark. Using the unique view adaption module, each network first identifies the best observation perspectives and then transforms the skeletons for end-to-end detection with the main classification network based on those viewpoints. Using the skeletons of different perspectives, the suggested view adaptive models can provide significantly more consistent virtual viewpoints. By removing views, the models allow networks to learn action-specific properties more efficiently. Furthermore, we developed a two-stream scheme (referred to as VA-fusion) that integrates the performance of two networks to obtain an improved prediction. Random rotation of skeletal sequences is used to avoid overfitting during training and improve the reliability of view adaption models. An extensive experiment demonstrates that our proposed view-adaptive networks outperform existing solutions on five challenging benchmarks.

Skeleton-based action recognition using graph convolutional networks (GCNs), which specify CNNs to a more flexible non-Euclidean frame, has shown outstanding results. However, many problems remain the same in the earlier GCN-based models. (I) All model layers and input data have the same graph structure. Given the GCN model’s hierarchy and the variety of action recognition input, this may not be appropriate. (II) Bone length and orientation are rarely studied because they are too helpful and different for human action recognition. This paper Article Title presents an extended multi-stream adaptive graph convolutional neu-ral network (EMS-AAGCN) for skeleton-based action recognition. The proposed model’s network topology can be trained uniformly or individually based on the input data. This approach is based on data, allowing the model to make graphs more flexible and quickly adapt to various datasets. In addition, a spatial-temporal channel attention module in the suggested adaptive graph convolutional layer allows it to focus more on joints, frames, and features. Furthermore, an enhanced multi-stream framework models joints,bones and their motion, improving recognition accuracy. Our method significances the state-of-the-art on the two massive datasets, NTU-RGBD and Kinetics-Skeleton.