I would like to learn and discuss about a new method for identification based on "i-vectors". It was first proposed for speaker identification. But now people are trying to apply it on "face identification", and many other problems of clustering.
Here is the original paper:
http://groups.csail.mit.edu/sls/publications/2011/Dehak_IEEE_May2011.pdf
We may also refer to this tutorial:
http://www.icsi.berkeley.edu/Speech/presentations/AFRL_ICSI_visit2_JFA_tutorial_icsitalk.pdf
Best,
Elie.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment