Optimal face reconstruction using training

In previous work [2] we considered the problem of image interpolation from an adaptive optimal recovery point of view. We showed how a training set S determines a quadratic signal class and how to use this signal class to perform image interpolation. In [2] the training set S was taken from the low resolution version of the image we were interpolating. In this paper we continue our discussion of the method presented in [2] by looking more closely at the training set S. In particular, we will show how a training set of high resolution images can give very good interpolation results through the use of the method [2].

• Type: Conference
• Year: 2002
• Where: ICIP
• Authors: D. Darian Muresan, Thomas W. Parks
• Keywords: ICIP 2002, Image Interpolation, Image Processing, Quadratic Signal Class, Training Set