A survey on image morphing

PhD Qualifying Examination


Title: "A survey on image morphing"

by

Miss Jing LIAO


Abstract:

Image morphing is a powerful tool for visual effects that consists of the 
fluid transformation of one digital image into another. It has been long 
and widely used in creating special effects for movie, television 
commercials, and music videos.

Most Image morphing algorithms share the following components: feature 
specification, warp generation and color interpolation. It firstly 
requires user to specify feature correspondences with pairs of feature 
primitives, e.g. mesh nodes, lines, curves or points. And then two input 
images are warped to intermediate states according to feature 
correspondences. Lastly ordinary color interpolation is simply applied to 
warped images, generating each in-between frame.

The ease of use of image morphing depends on feature specification, so we 
categorize existing image morphing algorithms into four categories 
according to user specified features: mesh-based, line-based, point-based 
and image content based. And representative algorithms, advantages and 
drawbacks of each category are described in this survey.


Date:			Monday, 17 June 2013

Time:                   3:30pm - 5:30pm

Venue:                  Room 3501
                        Lifts 25/26

Committee Members:	Dr. Pedro Sander (Supervisor)
 			Prof. Long Quan (Chairperson)
 			Dr. Huamin Qu
 			Dr. Chiew-Lan Ti


**** ALL are Welcome ****