An Enhanced Approach on Temporal Contrastive learning for Human Activity Recognition Using Deep Learning Technique

Human Activity Recognition systems develop important values in healthcare monitoring, interactive gaming, and surveillance. Temporal Graph Contrastive Learning has emerged as one of the powerful approaches towards deep-learning-based technique. Provisionally, TCNs offer a very strong framework that elicits temporal suggestions and subtleties in sequential data without the problematic and training challenges associated with RNNs. By using contrastive learning, TGCL learns more discriminative temporal features from complex dynamic activity data. In this way, robustness is enhanced, and accuracy is improved in systems of human activity recognition. the performance by allowing distinguishing activities that are pretty similar in nature. The capability of TGCL to incorporate spatial and transient aspects into a coherent learning experience allows it to emerge as a promising solution to advance the state of the art in progressive HAR, in the most complex and dynamic scenarios.