Closed lqql2012 closed 6 years ago
Hi, I'm sorry, but from your description it's not really possible to tell what you're trying to do or why it may go wrong. It sounds very much like it has nothing to do with eos though, we're training a variety of models successfully with it. I suggest that you follow a tutorial about PCA and get more familiar with the matter first, and maybe step through a simpler example.
i am try to create full head model by 20 obj file,first i get mean_face eigenvalues and eigenvectors,But there is no difference in the output face!!! and if i enlarge fitted_coeffs*60 ,get result as below
So i try to use mean_face eigenvalues and eigenvectors that from original data(sfm_shape_3448.bin extract 20 dimension ),and it work!!
By comparing the pca parameters, there may be problems with the processing of eigenvectors !But I don't know the correct way to deal with it
Looking forward to your answer!!