Nonnegative CCA for Audiovisual Source Separation

  title={Nonnegative CCA for Audiovisual Source Separation},
  author={Christian Sigg and Bruno Brand{\~a}o Fischer and Bjorn Ommer and Volker Roth and Joachim M. Buhmann},
  journal={2007 IEEE Workshop on Machine Learning for Signal Processing},
We present a method for finding correlated components in audio and video signals. The new technique is applied to the task of identifying sources in video and separating them in audio. The concept of canonical correlation analysis is reformulated such that it incorporates nonnegativity and sparsity constraints on the coefficients of projection directions. Nonnegativity ensures that projections are compatible with an interpretation as energy signals. Sparsity ensures that coefficient weight… CONTINUE READING
Highly Cited
This paper has 19 citations. REVIEW CITATIONS