Szczegóły publikacji
Opis bibliograficzny
Real-time multiview human body tracking using GPU-accelerated PSO / Bogusław Rymut, Bogdan KWOLEK // W: Parallel Processing and Applied Mathematics : 10th international conference, PPAM 2013 : Warsaw, Poland, September 8–11, 2013 : revised selected papers, Pt. 1 / eds. Roman Wyrzykowski, [et al.]. — Berlin ; Heidelberg : Springer-Verlag, cop. 2014. — (Lecture Notes in Computer Science ; ISSN 0302-9743 ; 8384). — ISBN: 978-3-642-55223-6; e-ISBN: 978-3-642-55224-3. — S. 458–468. — Bibliogr. s. 467–468, Abstr.
Autorzy (2)
- Rymut Bogusław
- AGHKwolek Bogdan
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 84695 |
|---|---|
| Data dodania do BaDAP | 2014-10-07 |
| DOI | 10.1007/978-3-642-55224-3_43 |
| Rok publikacji | 2014 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Konferencja | International Conference on Parallel Processing and Applied Mathematics 2013 |
| Czasopisma/serie | Lecture Notes in Computer Science, Theoretical Computer Science and General Issues |
Abstract
This paper presents our approach to 3D model-based human motion tracking using a GPU-accelerated particle swarm optimization. The tracking involves configuring the 3D human model in the pose described by each particle and then rasterizing it in each particle's 2D plane. In our implementation, we launch one independent thread for each column of each 2D plane. Such a parallel algorithm exhibits the level of parallelism that allows us to effectively utilize the GPU resources. Owing to such task decomposition the tracking of the full human body can be performed at rates of 15 frames per second. The GPU achieves an average speedup of 7.5 over the CPU. The speedup that achieves the GPU over CPU grows with the number of the particles. For marker-less motion capture system consisting of four calibrated and synchronized cameras, the efficiency comparisons were conducted on four CPU cores and four GTX GPUs on two cards.
