![Institutional Repository [SANDBOX]](images/dias.logo.png "Institutional Repository [SANDBOX]"){kind=logo}
Institutional Repository [SANDBOX]
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/CA480050-A068-47EC-8B2F-3F16E7F3327A | ||
| Year | 2022 | ||
| Type of Item | Conference Full Paper | ||
| License |
|
||
| Bibliographic Citation | D. Kastrinakis and E. G. M. Petrakis, “V2F: real time video segmentation with Apache Flink,” in Advances in Visual Computing, vol 13599, Lecture Notes in Computer Science, G. Bebis, B. Li, A. Yao, Y. Liu, Y. Duan, M. Lau, R. Khadka, A. Crisan, R. Chang, Eds., Cham, Switzerland: Springer, 2022, pp. 153–164, doi: 10.1007/978-3-031-20716-7_12. https://doi.org/10.1007/978-3-031-20716-7_12 | ||
| Appears in Collections |
V2F is a distributed video processing system for bounded (i.e., stored) and unbounded (i.e., continuous) or real time video streams. Apache Flink applies a series of operators in a pipeline to transform a video stream into shots. These operators are replicated to work in parallel on Flink-managed computing nodes. The V2F deployment of the standard twin-comparison video segmentation method is more than 7 times faster than its non-parallel (i.e., sequential) implementation.
Copyright © DIAS 2013
