期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2019
卷号：7
期号：12
页码：4401-4411
DOI：10.15680/IJIRCCE.2019. 0712001
出版社：S&S Publications
摘要：Stitching collection of images from a variety of viewpoints into a panoramic image is a common
function. But when extending this panoramic image to video, there are many challenges. Video stitching has to
deal with moving objects in the overlapped area which often cause problems of structural misalignment or ghost
effect. The goal of multi-view video stitching is to generate a panoramic video from multiple overlapping video
streams captured by relatively displaced cameras. The input to a multi-view video stitching program is multiple,
overlapping videos captured from different camera views which are together called a multi-view video, and the
output is a panoramic video of wider field of view constructed by merging and stitching the individual videos of
the input multi-view video. Feature extraction, Feature Matching, homography estimation, Stitching and
stabilization are the different steps needed to generate panorama video. Panoramic video stitching is one of the
common research area in computer vision, image/video processing and computer graphics and has many wide
applications.
关键词：feature extraction; feature matching; homography estimation; video Stitching; blending
