Guest Editor(s)
Professor Dr. Sahraoui Dhelim
Email: sahraoui.dhelim@dcu.ie
Affiliation: Faculty of Engineering and Computing, Dublin City University, Dublin, Ireland
Homepage:
Research Interests: artificial intelligence, data analytics, security and privacy, internet of things, ITS, quantum computing
Professor Dr. Anas Bilal
Email: 910288@hainnu.edu.cn
Affiliation: College of Information Science and Technology, Hainan Normal University, Haikou, 571158, China
Homepage:
Research Interests: artificial intelligence, computer vision, image processing, medical image analysis, pattern recognition, quantum computing, and optimization algorithms
Prof. Dr Raheem Sarwar
Email: r.sarwar@mmu.ac.uk
Affiliation: OTEHM, Manchester Metropolitan University, Manchester, United Kingdom
Homepage:
Research Interests: artificial intelligence, computer vision, image processing, medical image analysis, pattern recognition, quantum computing, and optimization algorithms
Summary
Action Recognition (AR) is a critical area of research that enables machines to understand human activities from multimodal data sources such as video, sensor data, and wearable devices. This special issue for CMC—Computers, Materials & Continua focuses on cutting-edge advancements in AR, with a strong emphasis on privacy, explainability, and optimization for real-world applications. We invite contributions that push the boundaries of AR by addressing the challenges of real-time processing, robustness in noisy environments, and ensuring privacy-preserving techniques.
Submissions are sought on innovative methods for action recognition that involve deep learning architectures, such as convolutional networks, recurrent networks, transformers, and graph-based models. We are particularly interested in methods that enhance the efficiency of AR algorithms, including real-time action recognition, optimization for computational resources (e.g., latency, energy, throughput), and techniques for increasing robustness and generalization across diverse datasets. Additionally, submissions should focus on providing explanations for model decisions through explainable AI approaches, ensuring that AR models are interpretable and trustworthy in practical applications.
Potential Topics (not limited to):
· Deep Learning for Action Recognition (e.g., CNNs, RNNs, Transformers)
· Real-Time Action Recognition and Optimization (Latency, Energy, Throughput)
· Privacy-Preserving Techniques for Action Recognition
· Multimodal Data Fusion for Action Recognition
· Explainable AI for Action Recognition Models
· Robust Action Recognition Under Adversarial Conditions
· Self-Supervised and Few-Shot Learning for Action Recognition
· Benchmarking and Datasets for Action Recognition
Keywords
action recognition, human activity recognition, deep learning, video-based action recognition, multimodal sensing, privacy-preserving AI, real-time recognition, explainable AI, adversarial robustness, optimization, self-supervised learning, few-shot learning, action recognition datasets.
Published Papers
Login
Register
Submit a Paper
Propose a Special lssue