Age and Gender Detection from scratch

• Convolutional Neural Network (CNN) implementation for detecting age and gender from facial images, built from scratch.
• CNN Architecture: Implemented using Convolutional Neural Networks combined with MaxPooling layers.
• Libraries: TensorFlow, Keras, OpenCV, Matplotlib, NumPy, pandas, and scikit-learn
• Dataset: UTKFace
• Activation Function: ReLU
• Optimizers: Adam for Age Regression and Sigmoid for Gender Classification
• Loss Function: Mean Squared Error(MSE)
• Metric: Mean Absolute Error(MAE)
• Accuracy: Achieved an overall accuracy of 89% on the test dataset.
• Results:

old.man.mp4

woman.mp4

child.mp4

colleagues.mp4

man.mp4