Parts-based segmentation with overlapping part models using Markov chain Monte Carlo

Authors:
Matthias Seise;Stephen J. McKenna;Ian W. Ricketts;Carlos A. Wigderowitz
Affiliations:
School of Computing, University of Dundee, UK;School of Computing, University of Dundee, UK;School of Computing, University of Dundee, UK;Orthopaedic & Trauma Surgery, Ninewells Hospital, Dundee, UK
Venue:
Image and Vision Computing
Year:
2009

Citing 7
Cited 0

Hands: a pattern theoretic study of biological shapes

Hands: a pattern theoretic study of biological shapes
Deformable Template Recognition of Multiple Occluded Objects

IEEE Transactions on Pattern Analysis and Machine Intelligence
Convergence assessment techniques for Markov chain Monte Carlo

Statistics and Computing
Identification of partly destroyed objects using deformable templates

Statistics and Computing
Hierarchical Shape Modeling for Automatic Face Localization

ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part II
Information Theory, Inference & Learning Algorithms

Information Theory, Inference & Learning Algorithms
Recognizing multiple overlapping objects in image: an optimal formulation

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

A probabilistic method is proposed for segmenting multiple objects that overlap or are in close proximity to one another. A likelihood function is formulated that explicitly models overlapping object appearance. Priors on global appearance and geometry (including shape) are learned from example images. Markov chain Monte Carlo methods are used to obtain samples from a posterior distribution over model parameters from which expectations can be estimated. The method is described in detail for the problem of segmenting femur and tibia in X-ray images. The result is a probabilistic segmentation that quantifies uncertainty, conditioned upon the model, so that measurements such as joint space can be made with associated uncertainty.