Parallel Gradient Domain Processing of Massive Images

Philip, Sujin; Summa, Brian; Bremer, Peer-Timo; Pascucci, Valerio

Parallel Gradient Domain Processing of Massive Images

Files

011-019.pdf (1.85 MB)

Date

2011

Authors

Philip, Sujin
Summa, Brian
Bremer, Peer-Timo
Pascucci, Valerio

Publisher

The Eurographics Association

Abstract

Gradient domain processing remains a particularly computationally expensive technique even for relatively small images. When images become massive in size, giga or terapixel, these problems become particularly troublesome and the best serial techniques take on the order of hours or days to compute a solution. In this paper, we provide a simple framework for the parallel gradient domain processing. Specifically, we provide a parallel out-of-core method for the seamless stitching of gigapixel panoramas in a parallel MPI environment. Unlike existing techniques, the framework provides both a straightforward implementation, maintains strict control over the required/allocated resources, and makes no assumptions on the speed of convergence to an acceptable image. Furthermore, the approach shows good weak/strong scaling from several to hundreds of cores and runs on a variety of hardware.

        @inproceedings{10.2312:EGPGV/EGPGV11/011-019
,
booktitle = {Eurographics Symposium on Parallel Graphics and Visualization
},
editor = {Torsten Kuhlen and Renato Pajarola and Kun Zhou
},
title = {{Parallel Gradient Domain Processing of Massive Images
}},
author = {Philip, Sujin and 
Summa, Brian and 
Bremer, Peer-Timo and 
Pascucci, Valerio
},
year = {2011
},
publisher = {The Eurographics Association
},
ISSN = {1727-348X
},
ISBN = {978-3-905674-32-3
},
DOI = {10.2312/EGPGV/EGPGV11/011-019
}
}