Results 1 to 3 of 3

Math Help - Combining PCA models

  1. #1
    Newbie
    Joined
    Mar 2014
    From
    New Zealand
    Posts
    5

    Combining PCA models

    Hello,

    I am building an active appearance model. I have a large training data set. Too big. :/

    My computer's RAM cannot handle a model of this size.

    If i make a series of smaller PCA models. Is there a way to combine a series of PCA models into one 'averaged' projection space?

    Each sample is an image, that is reshaped into a 1D vector.

    Vectors are stacked into one large training data matrix, where each row is one sample. We call OpenCV's PCA function and get the 1D eigenvectors and a 2D eigenvalues(1 row per sample) matrix. We are interested in the 1D eigenvector matrix.

    Somebody has suggested Structural Equation Modelling. I thought an average of the eigenvectors should work.... provided the components correspond throughout the series of models.

    Any clues anyone?
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Newbie
    Joined
    Mar 2014
    From
    New Zealand
    Posts
    5

    Re: Combining PCA models

    Here we have a solution, however i don't really understand it.

    http://www.cs.cf.ac.uk/Dave/Papers/Pami-EIgenspace.pdf
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Newbie
    Joined
    Mar 2014
    From
    New Zealand
    Posts
    5

    Re: Combining PCA models

    Anyone?
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. combining pdfs
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: October 18th 2010, 04:17 PM
  2. combining functions
    Posted in the Algebra Forum
    Replies: 3
    Last Post: April 26th 2008, 06:20 PM
  3. Combining log and ln?
    Posted in the Algebra Forum
    Replies: 7
    Last Post: January 13th 2008, 02:53 PM
  4. Combining Vectors
    Posted in the Geometry Forum
    Replies: 1
    Last Post: September 4th 2007, 12:03 PM
  5. Replies: 2
    Last Post: July 17th 2007, 06:11 AM

Search Tags


/mathhelpforum @mathhelpforum