Project information

• Project date: March 2024
• Tech used: Deep Learning, Machine Learning, Convolutional Neural Networks(CNN), Python
GitHub link

This project aims to develop a robust model for Human Activity Recognition (HAR) using deep learning techniques. Given the significance of HAR in various applications such as surveillance, healthcare, and human- computer interaction, our study employs convolutional neural networks (CNNs) and transfer learning models like MobileNetV2 and EfficientNetB7 to classify activities in images. Results demonstrate the potential of EfficientNetB7 in achieving higher accuracy, offering insights into model performance and future research directions.