he files in this package comprise the Matlab implementation of a

foreground segmentation algorithm based upon graph cuts, described in:



Better Foreground Segmentation Through Graph Cuts, N. Howe &

A. Deschamps. Tech report, http://arxiv.org/abs/cs.CV/0401017.



The file extractForeground.m contains a sample function that will

perform a complete foreground segmentation for static camera video.

It uses a number of parameters, which are described in the

documentation. An attempt was made to choose sensible default values,

but they may need to be adjusted for some videos.



More generally, the implementation in extractForeground.m is only a

suggestion of how the graph-based segmentation might be used. The

same approach can be applied with other (perhaps time-adaptive)

background models. The key step is at line 113 of the code, once the

deviation array has been created. The graph-cut segmentation can be

applied similarly to a deviation matrix computed in any other way.-he files in this package comprise the Matlab implementation of a

foreground segmentation algorithm based upon graph cuts, described in:



Better Foreground Segmentation Through Graph Cuts, N. Howe &

A. Deschamps.　Tech report, http://arxiv.org/abs/cs.CV/0401017.



The file extractForeground.m contains a sample function that will

perform a complete foreground segmentation for static camera video.

It uses a number of parameters, which are described in the

documentation.　An attempt was made to choose sensible default values,

but they may need to be adjusted for some videos.



More generally, the implementation in extractForeground.m is only a

suggestion of how the graph-based segmentation might be used.　The

same approach can be applied with other (perhaps time-adaptive)

background models.　The key step is at line 113 of the code, once the

deviation array has been created.　The graph-cut segmentation can be

applied similarly to a deviation matrix computed in any other way.