Principal geodesic analysis

Principal Geodesic Analysis (PGA) is a statistical method used for analyzing data that lies on a manifold, such as shapes, curves, or other geometric structures. This approach extends the traditional principal component analysis (PCA) to the context of Riemannian manifolds, which are spaces where the notion of distance and angles can vary in different directions. While PCA is effective for linear data in Euclidean spaces, PGA is designed to handle nonlinear data that resides on curved spaces.