Svd

Explainer breaks down how truncated SVD compresses image data using PCA math

A technical write-up walks through the relationship between Principal Component Analysis and Singular Value Decomposition, showing how a matrix can be factored into three components (U, Sigma, V) to capture the most significant variation in data. It uses a 1024x1024 grayscale image of the moon as a worked example, treating the image as a set of vectors that can be decomposed and reduced via truncated SVD.