covariance for cell array

I was just started applying Principal Component analysis on 3D face data set. I arrange the data in a cell. I try to find the covariance using the below code:

   B = cat(3, mydata{:});
   meanface = mean(B,3);
   C = B - repmat(meanface, [1 1 6]);
   Cov = C * permute(C,[2 3 1]);

Since transpose can't be used for 3d array. However i got an error : Error using ==> mtimes. Input arguments must be 2-D. How should i fix this? Thank you.