Parallelizing tracking algorithms
Material type:
ArticleDescription: 1 archivo (63,3 KB)Subject(s): Online resources: Summary: In several applications, the trajectory of an entity, a feature or an object has to be tracked over a sequence of image frames. When the processing is to be performed in real time, there are important constraints leading to the parallelization of tracking algorithms. This paper presents the results of a concrete implementation, which deals with the particular case of simple objects moving in an context reachable by the vision element (video camera). The steps involved in the solution development are detailed, specially in relation to their parallelization by using a computer heterogeneous network and MPI (Message Passing Interface) support. Finally, an analysis of the different algorithms behavior is carried out together with the obtained results assessment, which allows knowing the performed parallelization efficiency, and determining under which conditions this solution turns out to be the best one.
| Item type | Home library | Collection | Call number | URL | Status | Date due | Barcode | |
|---|---|---|---|---|---|---|---|---|
Capítulo de libro
|
Biblioteca de la Facultad de Informática | Biblioteca digital | A0597 (Browse shelf(Opens below)) | Link to resource | No corresponde |
Formato de archivo: PDF. -- Este documento es producción intelectual de la Facultad de Informática - UNLP (Colección BIPA/Biblioteca)
In several applications, the trajectory of an entity, a feature or an object has to be tracked over a sequence of image frames. When the processing is to be performed in real time, there are important constraints leading to the parallelization of tracking algorithms. This paper presents the results of a concrete implementation, which deals with the particular case of simple objects moving in an context reachable by the vision element (video camera). The steps involved in the solution development are detailed, specially in relation to their parallelization by using a computer heterogeneous network and MPI (Message Passing Interface) support. Finally, an analysis of the different algorithms behavior is carried out together with the obtained results assessment, which allows knowing the performed parallelization efficiency, and determining under which conditions this solution turns out to be the best one.
Journal of Computer Science & Technology, 1(6)