PCA-MS: Variational Multi-Phase Segmentation using High-Dimensional Local Features

A variational multi-phase segmentation framework based on the Mumford-Shah energy, combined with PCA-based dimension reduction is used to segment color or gray-value images into regions of different structure identified by high-dimensional features, such as local spectral histograms (Texture) or localized Fourier transforms (Crystals).


CONTENTS: C++ source code reproducing texture and crystal segmentation results presented at the IEEE Winter Conference on Applications of Computer Vision (WACV 2016) - see reference below.

|- quocmesh/                                        source code
|+ - fini...jects/highDi...tation/applyTxtMerge.m   MATLAB script for TxtMerge post-processing
|
|- quocGCC/                                         compilation folder
|+ - go.sh                                          bash script for CMake (see README.txt)
|+ - finishedProjects/highDimFeatureSegmentation    executables (created during compilation)
|
|- LICENSE.txt                                      Common Development and Distribution License
|- README.txt                                       intructions for compilation and execution

INSTRUCTIONS: Please refer to the accompanying README.txt


DATA & RESULTS: http://nmevenkamp.github.io/pcams-data-wacv2016


LICENSE: PCA-MS is distributed under the terms of the Common Development and Distribution License.


REFERENCE: Mevenkamp, N., and Berkels, B. Variational Multi-Phase Segmentation using High-Dimensional Local Features. Applications of Computer Vision (WACV), 2016 IEEE Winter Conference on, 2016, (accepted)


CONTACT:
Niklas Mevenkamp
Aachen Institute for Advanced Study in Computational Engineering Science
RWTH Aachen University
mevenkamp@aices.rwth-aachen.de
http://www.aices.rwth-aachen.de/people/mevenkamp