Details
| Originalsprache | Englisch |
|---|---|
| Seiten | 139-147 |
| Seitenumfang | 9 |
| Publikationsstatus | Veröffentlicht - 19 März 2014 |
| Extern publiziert | Ja |
| Veranstaltung | 5th ACM Multimedia Systems Conference, MMSys 2014 - Singapore, Singapur Dauer: 19 März 2014 → 21 März 2014 |
Konferenz
| Konferenz | 5th ACM Multimedia Systems Conference, MMSys 2014 |
|---|---|
| Land/Gebiet | Singapur |
| Ort | Singapore |
| Zeitraum | 19 März 2014 → 21 März 2014 |
Abstract
In this paper, we present a fast parallel algorithm for the retargeting of videos. It combines seam carving and cropping and is aimed for real-time adaptation of video streams. The basic idea is to first find an optimal cropping path over the whole sequence with the target size. Then, the borders are slightly extended to be reduced again by seam carving on a frame-by-frame basis. This allows the algorithm to get more important content into the cropping window as it is also able to remove pixels from within the window. In contrast to the previous SeamCrop algorithm, the presented technique is optimized for parallel processes and a CUDA GPU implementation. In comparison, the computation time of our GPU algorithm is 10:5 times faster (on a 960 × 540 video with a retarget factor of 25%) than the already efficient CPU implementation.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Computergrafik und computergestütztes Design
- Informatik (insg.)
- Maschinelles Sehen und Mustererkennung
- Informatik (insg.)
- Mensch-Maschine-Interaktion
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
2014. 139-147 Beitrag in 5th ACM Multimedia Systems Conference, MMSys 2014, Singapore, Singapur.
Publikation: Konferenzbeitrag › Paper › Forschung › Peer-Review
}
TY - CONF
T1 - GPU video retargeting with parallelized SeamCrop
AU - Kiess, Johannes
AU - Gritzner, Daniel
AU - Guthier, Benjamin
AU - Kopf, Stephan
AU - Effelsberg, Wolfgang
PY - 2014/3/19
Y1 - 2014/3/19
N2 - In this paper, we present a fast parallel algorithm for the retargeting of videos. It combines seam carving and cropping and is aimed for real-time adaptation of video streams. The basic idea is to first find an optimal cropping path over the whole sequence with the target size. Then, the borders are slightly extended to be reduced again by seam carving on a frame-by-frame basis. This allows the algorithm to get more important content into the cropping window as it is also able to remove pixels from within the window. In contrast to the previous SeamCrop algorithm, the presented technique is optimized for parallel processes and a CUDA GPU implementation. In comparison, the computation time of our GPU algorithm is 10:5 times faster (on a 960 × 540 video with a retarget factor of 25%) than the already efficient CPU implementation.
AB - In this paper, we present a fast parallel algorithm for the retargeting of videos. It combines seam carving and cropping and is aimed for real-time adaptation of video streams. The basic idea is to first find an optimal cropping path over the whole sequence with the target size. Then, the borders are slightly extended to be reduced again by seam carving on a frame-by-frame basis. This allows the algorithm to get more important content into the cropping window as it is also able to remove pixels from within the window. In contrast to the previous SeamCrop algorithm, the presented technique is optimized for parallel processes and a CUDA GPU implementation. In comparison, the computation time of our GPU algorithm is 10:5 times faster (on a 960 × 540 video with a retarget factor of 25%) than the already efficient CPU implementation.
KW - Cropping
KW - GPU
KW - Seam carving
KW - SeamCrop
KW - Video resizing
KW - Video retargeting
UR - http://www.scopus.com/inward/record.url?scp=84899008791&partnerID=8YFLogxK
U2 - 10.1145/2557642.2557648
DO - 10.1145/2557642.2557648
M3 - Paper
AN - SCOPUS:84899008791
SP - 139
EP - 147
T2 - 5th ACM Multimedia Systems Conference, MMSys 2014
Y2 - 19 March 2014 through 21 March 2014
ER -