Deep Learning

Age & Gender Prediction + Food Recommendation

A deep learning project that uses a Convolutional Neural Network to predict a person's age and gender from facial images, built on TensorFlow and Keras with OpenCV handling face detection in the preprocessing pipeline. Gender is predicted as a classification task while age is estimated via regression. As a bonus feature, a rule-based food recommendation system suggests nutritionally appropriate food once a face is classified into an age group — combining computer vision with a practical recommendation layer.

Model Accuracy

90%

Model Type

CNN (Age + Gender)

Category

Deep Learning

Tech Stack

6 Tools

Key Features

  • CNN-based facial analysis predicting age and gender simultaneously
  • OpenCV face-detection preprocessing pipeline
  • Transfer learning support for faster training, plus data augmentation
  • Bonus rule-based food recommendation engine keyed to predicted age group
  • Training curves, sample predictions & confusion-matrix visualizations
  • Reproducible end-to-end Jupyter Notebook workflow

Technologies

TensorFlowKerasOpenCVNumPyPandasMatplotlib

Performance Metrics

Gender Accuracy

~90%

Age Estimation MAE

~5 yrs

Face Detection

Real-time

Bonus Module

Food Reco

Interested in this project?