Improved vectorization of OpenCV algorithms for RISC-V CPUs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Volokitin, V. D., Vasiliev, E. P., Kozinov, E. A., Kustikova, V. D., Liniov, A. V., Rodimkov, Y. A., Sysoyev, A. V., Meyerov, I. B.
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914801305780224
author Volokitin, V. D.
Vasiliev, E. P.
Kozinov, E. A.
Kustikova, V. D.
Liniov, A. V.
Rodimkov, Y. A.
Sysoyev, A. V.
Meyerov, I. B.
author_facet Volokitin, V. D.
Vasiliev, E. P.
Kozinov, E. A.
Kustikova, V. D.
Liniov, A. V.
Rodimkov, Y. A.
Sysoyev, A. V.
Meyerov, I. B.
contents The development of an open and free RISC-V architecture is of great interest for a wide range of areas, including high-performance computing and numerical simulation in mathematics, physics, chemistry and other problem domains. In this paper, we discuss the possibilities of accelerating computations on available RISC-V processors by improving the vectorization of several computer vision and machine learning algorithms in the widely used OpenCV library. It is shown that improved vectorization speeds up computations on existing prototypes of RISC-V devices by tens of percent.
format Preprint
id arxiv_https___arxiv_org_abs_2311_12808
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Improved vectorization of OpenCV algorithms for RISC-V CPUs
Volokitin, V. D.
Vasiliev, E. P.
Kozinov, E. A.
Kustikova, V. D.
Liniov, A. V.
Rodimkov, Y. A.
Sysoyev, A. V.
Meyerov, I. B.
Distributed, Parallel, and Cluster Computing
Computer Vision and Pattern Recognition
Performance
The development of an open and free RISC-V architecture is of great interest for a wide range of areas, including high-performance computing and numerical simulation in mathematics, physics, chemistry and other problem domains. In this paper, we discuss the possibilities of accelerating computations on available RISC-V processors by improving the vectorization of several computer vision and machine learning algorithms in the widely used OpenCV library. It is shown that improved vectorization speeds up computations on existing prototypes of RISC-V devices by tens of percent.
title Improved vectorization of OpenCV algorithms for RISC-V CPUs
topic Distributed, Parallel, and Cluster Computing
Computer Vision and Pattern Recognition
Performance
url https://arxiv.org/abs/2311.12808