Piecewise constant nonnegative matrix factorization

Nicolas Seichepine, Slim Essid, Cédric Févotte, Olivier Cappé

Abstract

In this paper we propose a non-negative matrix factorization (NMF) model with piecewise-constant activation coefficients. This structure is enforced using a total variation penalty on the rows of the activation matrix. The resulting optimization problem is solved with a majorization-minimization procedure. The proposed algorithm is well suited to analyze data explained by underlying piecewise-constant sequences of states. Its properties are first illustrated using synthetic data. We then use it to solve a video structuring problem that involves both segmentation and clustering tasks. An improvement over a state-of-the-art temporally smoothed NMF algorithm of both clustering and segmentation quality measures is observed.
BibTeX
@inproceedings{seichepine2014piecewise,
  author = {Nicolas Seichepine and Slim Essid and Cédric Févotte and Olivier Cappé},
  title = {{Piecewise constant nonnegative matrix factorization}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2014},
  doi = {10.1109/icassp.2014.6854901},
}

Loading related content...

Figure