Visual surveillance using deep reinforcement learning

Keong Hun Choi, Jong Eun Ha

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Visual surveillance aims a robust detection of foreground objects, and traditional algorithms usually use a background model image. A current is compared with the background model image. In this paper, we present a visual surveillance algorithm, which determines the parameters in Vibe using deep reinforcement learning. We apply DQN to determine three parameters in Vibe algorithm. We present a policy model which is composed of encoder and decoder type network. Experimental results shows the feasibility of the presented algorithm.

Original languageEnglish
Title of host publication2020 20th International Conference on Control, Automation and Systems, ICCAS 2020
PublisherIEEE Computer Society
Pages289-291
Number of pages3
ISBN (Electronic)9788993215205
DOIs
StatePublished - 13 Oct 2020
Event20th International Conference on Control, Automation and Systems, ICCAS 2020 - Busan, Korea, Republic of
Duration: 13 Oct 202016 Oct 2020

Publication series

NameInternational Conference on Control, Automation and Systems
Volume2020-October
ISSN (Print)1598-7833

Conference

Conference20th International Conference on Control, Automation and Systems, ICCAS 2020
Country/TerritoryKorea, Republic of
CityBusan
Period13/10/2016/10/20

Keywords

  • BGS
  • Deep learning
  • GAN
  • Segmentation
  • Visual surveillance

Fingerprint

Dive into the research topics of 'Visual surveillance using deep reinforcement learning'. Together they form a unique fingerprint.

Cite this