.....
This project predicts a person’s age based on their facial image using a Convolutional Neural Network (CNN). It includes training the model on the UTKFace dataset and deploying it in a real-time webcam-based application.
- Predicts age from facial images using CNN
- Trained on the UTKFace dataset (20K+ images)
- Real-time age detection via webcam using OpenCV
- Age range classification for clearer output (e.g., "10–20", "28–45")
- Name: UTKFace Dataset
- Size: ~20,000 facial images
- Labels: Age (0–116), Gender, Ethnicity (only Age used here)
- Format:
age_gender_ethnicity_date.jpg(age is extracted from the filename)
The model is a CNN built with TensorFlow/Keras: