|
Abstract
|
|
|---|---|
| Shot boundary detection is the ?rst step to be performed on any Content- based Video Retrieval system. The performance of such techniques must be eval- uated from a computational point of view since this is a very high demanding task and thus optimization strategies must be seeked. This paper proposes scal- able and portable implementations that grant low response times over two alter- native parallel architectures: a shared-memory symmetric multiprocessor and a Beowulf cluster. Two alternative parallel programing paradigms have been eval- uated, shared-memory and message passing, implementing several strategies for video segmentation and data access and analyzing load balancing issues. | |
|
International
|
Si |
|
Congress
|
11th International Conference on Computational and Mathematical Methods in Science and Engineering, CMMSE 2011 |
|
|
960 |
|
Place
|
Alicante, España |
|
Reviewers
|
Si |
|
ISBN/ISSN
|
978-84-614-6167-7 |
|
|
|
|
Start Date
|
26/06/2011 |
|
End Date
|
30/06/2011 |
|
From page
|
1 |
|
To page
|
12 |
|
|
Proc of the 11th International Conference on Computational and Mathematical Methods in Science and Engineering, CMMSE 2011 |