Learning to see
Classifying images with features and nearest neighbours, convolutional neural networks, and detection and segmentation with IoU, NMS and average precision.
About 5 hours of study
Learning objectives
- Classify images and measure accuracy on test data
- Compute CNN shapes, parameters and feature maps
- Score detections and masks with IoU, AP and Dice
Lessons
- 1
Classifying images with features and nearest neighbours Intermediate
Teach a program to name what is in a picture. Treat images as vectors, classify them with k-nearest neighbours, see why raw pixels fail when objects move, and build HOG-style features from gradient orientations that make classification robust.
- 2
Convolutional neural networks for images Advanced
Let the network learn its own kernels. Follow an image through convolution layers with many channels, ReLU and max pooling to a classifier, compute output shapes and parameter counts, and see why weight sharing and pooling suit images.
- 3
Object detection metrics and segmentation Advanced
Go from "what is in the image" to "where is it". Measure box overlap with IoU, remove duplicate detections with non-maximum suppression, score a detector with precision, recall and average precision, and evaluate segmentation masks with IoU and Dice.
Module practice
- Computer vision: Classify by the nearest imageEasy
- Computer vision: k nearest neighbours with a voteMedium
- Computer vision: A histogram of gradient orientationsHard
- Computer vision: Shapes and parameters through a CNNEasy
- Computer vision: Convolution, ReLU and max poolingMedium
- Computer vision: A multi-channel convolution layerHard
- Computer vision: Intersection over union of two boxesEasy
- Computer vision: Non-maximum suppressionMedium
- Computer vision: Detector precision, recall and APHard