By Anjelica Joy M. Labrador
On Friday, 01 September 2006 (10:00 am) at ICS LH-4

Abstract

Image stitching is a common practice in the generation of panoramic images. The aim of a stitching algorithm is to produce a visually plausible mosaic with two sought-after properties: (1) the mosaic should be as similar as possible to the input images and (2) the seam between the stitched images should be invisible.

Several formal cost functions for the evaluation of the quality of stitching were introduced in order to identify and obtain the best possible stitching method. The similarity of the input images and the visibility of the seam were defined in the gradient domain in these cost functions, minimizing the disturbing edges along the seam.

The approached was demonstrated in various applications, including generation of panoramic images, object blending, and removal of compression artifacts. Comparisons with other existing methods showed the benefits of optimizing the measures in the gradient domain.

Author Biography

Anjelica Joy M. Labrador is a senior undergraduate student of the Institute
of Computer Science. She is currently conducting a study on image stitching.