GLOMO: MATLAB Toolbox for learning layers from video

GLOMO (Greedy Learning Of Multiple Objects) is a MATLAB Toolbox for learning moving layers (i.e. 2D appearance models of objects or object's parts) from a video sequence. The theoretical foundations of the implemented algorithm can be found in Titsias and Williams (2004), which builds upon the greedy algorithm of Williams and Titsias (2004). A complete description of the algorithm can be also found in my PhD thesis (Titsias, 2005).

Download

The software consists of a set of MATLAB functions that can be downloaded as a compressed tar-file; click here. The software is accompanied with demos (see below) and documentation.

Demos

• demo1.m: This applies the algorithm to the Frey-Jojic sequence. To run this demo you have to download the file freyjojic.tar.gz and decompress it inside the directory GLOMO/videos/frey_jojic/


• demo2.m: This applies the algorithm to the panorama sequence. To run this demo you have to download the file panorama.tar.gz and decompress it inside the directory GLOMO/videos/panorama/


• demo3.m: This applies the algorithm to the arms-torso sequence. To run this demo you have to download the file armstorso.tar.gz and decompress it inside the directory GLOMO/videos/arms-torso/


• demo4.m: To run this demo you need to download the Frey-Jojic sequence as explained in the demo1.m case.

Acknowledgements: We thank Nebojsa Jojic and Brendan Frey for poviding the Fey-Jojic and panaroma sequences.

References