Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ South Ural State Uni...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
addClaim

This Research product is the result of merged Research products in OpenAIRE.

You have already added 0 works in your ORCID record related to the merged Research product.

Подходы к оптимизации и распараллеливанию вычислений в задаче детектирования объектов разных классов на изображении

Authors: Kozinov, E. A.; Kustikova, V. D.; Meyerov, I. B.; Polovinkin, A. N.; Sidnev, A. A.;

Подходы к оптимизации и распараллеливанию вычислений в задаче детектирования объектов разных классов на изображении

Abstract

Рассматривается задача детектирования объектов разных классов на статических изображениях: фотографиях или отдельных кадрах видеопотока. Описывается схема решения данной задачи с использованием алгоритма Latent SVM. Используется известный подход к ускорению вычислений - построение каскада классификаторов. Описывается вычислительная схема решения задачи детектирования с помощью каскадного Latent SVM. Обсуждаются проблемы распараллеливания и оптимизации времени поиска объектов одного класса на изображении. Проводится анализ вариантов решения указанных проблем. Выделяются наиболее трудоемкие участки реализаций, рассматриваются различные схемы распараллеливания, оцениваются их преимущества и недостатки. Приводятся результаты вычислительных экспериментов на базе изображений PASCAL Visual Object Challenge 2007, дается их анализ, а также формулируются выводы и планы по дальнейшему развитию. This paper considers the problem of object detection in static images. We describe a state-of-the-art method based on Latent SVM algorithm. A well-known approach to speed up calculations, the construction of cascade classifiers, is used. We describe a computational scheme that uses cascade modification of the original Latent SVM algorithm The issues of parallelization and performance optimization are discussed. We analyze the most timeconsuming parts of implementation, consider several parallelization schemes and aspects of their performance. The results of numerical experiments on PASCAL Visual Object Challenge 2007 image dataset are given. E.A. Kozinov, Nizhny Novgorod State University (Nizhny Novgorod, Russian Federation), V.D. Kustikova, Nizhny Novgorod State University (Nizhny Novgorod, Russian Federation), I.B. Meyerov, Nizhny Novgorod State University (Nizhny Novgorod, Russian Federation), A.N. Polovinkin, Nizhny Novgorod State University (Nizhny Novgorod, Russian Federation), A.A. Sidnev, Nizhny Novgorod State University (Nizhny Novgorod, Russian Federation)

Keywords

algorithm Latent SVM, каскадный классификатор, УДК 519.6, parallelization, cascade classifier, детектирование объектов, УДК 004.021, распараллеливание, object detection, алгоритм Latent SVM

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green