Research Article
Human Activity Recognition Using Convolutional Neural Networks in Multimedia Event Detection
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357987, author={Koushiram N and Rupendra Praveen Kumar and Kaavya Kanagraj}, title={Human Activity Recognition Using Convolutional Neural Networks in Multimedia Event Detection}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part II}, publisher={EAI}, proceedings_a={ICITSM PART II}, year={2025}, month={10}, keywords={human activity recognition sports analytics convolution neural network dataset deep learning models real-time activity}, doi={10.4108/eai.28-4-2025.2357987} }- Koushiram N
Rupendra Praveen Kumar
Kaavya Kanagraj
Year: 2025
Human Activity Recognition Using Convolutional Neural Networks in Multimedia Event Detection
ICITSM PART II
EAI
DOI: 10.4108/eai.28-4-2025.2357987
Abstract
Human Activity Recognition (HAR) serves as an essential element of AI and ML which plays an important role in healthcare supervision and sports monitoring and security systems development and smart environment performance. This research evaluates the application of Convolutional Neural Networks (CNNs) for Human Activity Recognition (HAR) since they can analyze both spatial and temporal elements from raw data inputs automatically. CNNs achieve high accuracy in action detection when they receive training from labeled sensor or video datasets to recognize strolling, running, resting and skipping behaviors. The proposed method demonstrates advantages for real-time processing capabilities along with ease of deployment across various contexts which make it selectable for real-world implementations. Research findings indicate that activity recognition applications benefit immensely from CNNs through their precise analysis along with their resistance to failures and high-speed operation.
