Human Action Image Classification

This project leverages deep learning and transfer learning techniques to classify human actions in images using pre-trained Convolutional Neural Network (CNN) models such as ResNet and VGG16. The models were fine-tuned on a human action image dataset to accurately recognize different types of actions.

📌 Project Objectives

🧠 Apply transfer learning with ResNet and VGG16 for image classification

🖼️ Classify images based on human actions (e.g., running, jumping, sitting, etc.)

📊 Compare models using evaluation metrics like accuracy, precision, recall, and F1-score

🛠️ Technologies Used

Python

TensorFlow / Keras

NumPy, Pandas

Matplotlib, Seaborn

Scikit-learn

Tkinter