Image Morphing Based on Mutual Information and Optimal Mass Transport (TA-P7)

Author(s) : Lei Zhu (Georgia Institute of Technology, USA) Yan Yang (Georgia Institute of Technology, USA) Allen Tannenbaum (Georgia Institute of Technology, USA) Steven Haker (Brigham and Women's Hospital and Harvard Medical School, USA)

Abstract : Time domain image interpolation, or image morphing, refers to a class of techniques for generating a series of smoothly changing intermediate images between two given related images. In this note, we present a novel approach based on the theory of optimal mass transport, using mutual information (MI) as the similarity measurement. The potential applications also include image registration, compression and coding.

Menu