ENHANCING HUMAN ACTIVITY RECOGNITION PERFORMANCE USING DEEP LEARNING TECHNIQUES

Abstract

Numerous astounding miracles that will improve our lives have been inspired by the advancement of technology. Computer vision, image detection, and facial recognition are innovative techniques that also make it possible to recognize human behavior. Activity tracking eliminates the need for manual entry and enables forecasting and analysis of human behavior, exposing hitherto hidden benefits. Here, we've explored many approaches to carrying out human activity recognition and contrasted them in terms of benefits, effectiveness, accuracy, methodology, datasets, and constraints. Along with the difficulties and the approach employed, we also covered several areas where this could be useful, including augmented reality, security, and the healthcare industry.