Bayesian clustering of shapes of curves

DOI10.1016/J.JSPI.2015.04.007zbMATH Open1394.62088arXiv1504.00377OpenAlexW2963895809MaRDI QIDQ899551FDOQ899551

Authors: Zhengwu Zhang, Debdeep Pati, Anuj Srivastava 

Publication date: 29 December 2015

Published in: Journal of Statistical Planning and Inference (Search for Journal in Brave)

Abstract: Unsupervised clustering of curves according to their shapes is an important problem with broad scientific applications. The existing model-based clustering techniques either rely on simple probability models (e.g., Gaussian) that are not generally valid for shape analysis or assume the number of clusters. We develop an efficient Bayesian method to cluster curve data using an elastic shape metric that is based on joint registration and comparison of shapes of curves. The elastic-inner product matrix obtained from the data is modeled using a Wishart distribution whose parameters are assigned carefully chosen prior distributions to allow for automatic inference on the number of clusters. Posterior is sampled through an efficient Markov chain Monte Carlo procedure based on the Chinese restaurant process to infer (1) the posterior distribution on the number of clusters, and (2) clustering configuration of shapes. This method is demonstrated on a variety of synthetic data and real data examples on protein structure analysis, cell shape analysis in microscopy images, and clustering of shaped from MPEG7 database.

Full work available at URL: https://arxiv.org/abs/1504.00377

Recommendations

• Bayesian shape clustering
• Unsupervised Bayesian clustering for functional data
• \(k\)-mean alignment for curve clustering
• Functional Clustering by Bayesian Wavelet Methods
• Applications of shape-distance metric to clustering shape-databases

zbMATH Keywords

clusteringWishart distributionChinese restaurant processshapes of curves

Mathematics Subject Classification ID

Bayesian inference (62F15) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Cites Work

• Generative models for functional data using phase and amplitude separation
• Model-Based Clustering, Discriminant Analysis, and Density Estimation
• How Many Clusters? Which Clustering Method? Answers Via Model-Based Cluster Analysis
• Model-Based Gaussian and Non-Gaussian Clustering
• Statistical modeling of curves using shapes and related features
• Prior distributions on spaces of probability measures
• A Bayesian analysis of some nonparametric problems
• Title not available (Why is that?)
• Title not available (Why is that?)
• Combinatorial stochastic processes. Ecole d'Eté de Probabilités de Saint-Flour XXXII -- 2002.
• Inconsistency of Pitman-Yor process mixtures for the number of components
• Projection-based curve clustering
• Title not available (Why is that?)
• A stratificational overlapping cluster scheme

Cited In (14)

• Phase-Amplitude Separation and Modeling of Spherical Trajectories
• Bayesian cluster analysis for registration and clustering homogeneous subgroups in multidimensional functional data
• Bayesian spatial homogeneity pursuit of functional data: an application to the U.S. income distribution
• Comparing model-based versus \(k\)-means clustering for the planar shapes
• Unsupervised Bayesian clustering for functional data
• Predictions based on the clustering of heterogeneous functions via shape and subject-specific covariates
• Elastic analysis of irregularly or sparsely sampled curves
• Deformation analysis in tunnels through curve clustering
• Fast and efficient Bayesian semi-parametric curve-fitting and clustering in massive data
• Depth for curve data and applications
• CATS
• Unsupervised Curve Clustering using B‐Splines
• Shape modeling with spline partitions
• Bayesian shape clustering

Uses Software

• mclust
• fdasrvf

This page was built for publication: Bayesian clustering of shapes of curves

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q899551)