/mcg

Multiscale Combinatorial Grouping - Object Proposals and Segmentation

Primary LanguageMATLABOtherNOASSERTION

For faster (0.8s/image) and improved hierarchical segmentation and proposals see "Convolutional Oriented Boundaries", ECCV 2016: http://www.vision.ee.ethz.ch/~cvlsegmentation/cob/

Multiscale Combinatorial Grouping

MCG is the package code that implements the algorithms presented in:

  • Pont-Tuset J, Arbelaez P, Barron J, Marques F, Malik J, "Multiscale Combinatorial Grouping for Image Segmentation and Object Proposal Generation," TPAMI 2016.

  • Arbelaez P, Pont-Tuset J, Barron J, Marques F, Malik J, "Multiscale Combinatorial Grouping," CVPR 2014.

Please consider citing the papers if you use this code.

More info in the project page: http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/mcg/

CAUTION

The 'full' package is not completely up to date, use it only if you need to re-train the Pareto front or the ranking in MCG, then use 'pre-trained' to do the actual computations.

'Full' and 'Pre-Trained' packages not compatible with Windows.