Title here
Summary here
Module · 6 lessons
Computer Vision with CNNs
Process images with convolutional neural networks, modern architectures, and transfer learning.
At a glance
Level
Advanced
Lessons
6 lessons
Time to complete
2–3 weeks
Cost
Free forever · no sign-up
Convolutional neural networks power modern computer vision. This module explains convolution and pooling, walks through influential CNN architectures, and shows how transfer learning lets you reuse powerful pretrained models — applied to a real medical imaging project.
Lessons in this module
1
Introduction to Convolutional Neural Networks
Understand why convolutional neural networks dominate computer vision, using the real Fashion-MNIST dataset
2
CNN Architecture
Build, compile, and train a convolutional neural network in Keras on the real Fashion-MNIST dataset
3
Regularization in Deep Learning
Fight overfitting in convolutional neural networks with dropout and data augmentation, using the real Fashion-MNIST dataset and Keras
4
Advanced CNN Architectures
Go deeper with convolutional networks: stack more blocks, add batch normalization and global average pooling, and learn the ideas behind VGG and ResNet
5
Transfer Learning
Reuse a model pretrained on ImageNet as a frozen feature extractor and train a small classifier head on Fashion-MNIST with far less data
6
Guided Project: Detecting Pneumonia from X-Ray Images
Build an end-to-end medical-imaging CNN on real chest X-rays from the PneumoniaMNIST dataset and learn why recall matters most in screening
Achievement
Complete all 6 lessons to finish the Computer Vision with CNNs module.