Accurate Human Localization for Automatic Labelling of Human from Fisheye Images
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Van Pha Than | - |
dc.contributor.author | Thanh Binh Nguyen | - |
dc.contributor.author | 정선태 | - |
dc.date.available | 2018-05-09T00:41:51Z | - |
dc.date.created | 2018-04-17 | - |
dc.date.issued | 2017-05 | - |
dc.identifier.issn | 1229-7771 | - |
dc.identifier.uri | http://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/6717 | - |
dc.description.abstract | Deep learning networks like Convolutional Neural Networks (CNNs) show successful performances in many computer vision applications such as image classification, object detection, and so on. For implementation of deep learning networks in embedded system with limited processing power and memory, deep learning network may need to be simplified. However, simplified deep learning network cannot learn every possible scene. One realistic strategy for embedded deep learning network is to construct a simplified deep learning network model optimized for the scene images of the installation place. Then, automatic training will be necessitated for commercialization. In this paper, as an intermediate step toward automatic training under fisheye camera environments, we study more precise human localization in fisheye images, and propose an accurate human localization method, Automatic Ground- Truth Labelling Method (AGTLM). AGTLM first localizes candidate human object bounding boxes by utilizing GoogLeNet-LSTM approach, and after reassurance process by GoogLeNet-based CNN network, finally refines them more correctly and precisely(tightly) by applying saliency object detection technique. The performance improvement of the proposed human localization method, AGTLM with respect to accuracy and tightness is shown through several experiments. | - |
dc.language | 영어 | - |
dc.language.iso | en | - |
dc.publisher | 한국멀티미디어학회 | - |
dc.relation.isPartOf | 멀티미디어학회논문지 | - |
dc.subject | Human Localization | - |
dc.subject | Fisheye Camera | - |
dc.subject | CNN (Convolutional Neural Networks) | - |
dc.subject | GoogLeNet | - |
dc.subject | Long Short Term Memory | - |
dc.subject | Saliency Detection | - |
dc.title | Accurate Human Localization for Automatic Labelling of Human from Fisheye Images | - |
dc.type | Article | - |
dc.identifier.doi | 10.9717/kmms.2017.20.5.769 | - |
dc.type.rims | ART | - |
dc.identifier.bibliographicCitation | 멀티미디어학회논문지, v.20, no.5, pp.769 - 781 | - |
dc.identifier.kciid | ART002225753 | - |
dc.description.journalClass | 2 | - |
dc.citation.endPage | 781 | - |
dc.citation.number | 5 | - |
dc.citation.startPage | 769 | - |
dc.citation.title | 멀티미디어학회논문지 | - |
dc.citation.volume | 20 | - |
dc.contributor.affiliatedAuthor | 정선태 | - |
dc.identifier.url | https://www.kci.go.kr/kciportal/ci/sereArticleSearch/ciSereArtiView.kci?sereArticleSearchBean.artiId=ART002225753 | - |
dc.description.isOpenAccess | N | - |
dc.subject.keywordAuthor | Human Localization | - |
dc.subject.keywordAuthor | Fisheye Camera | - |
dc.subject.keywordAuthor | CNN (Convolutional Neural Networks) | - |
dc.subject.keywordAuthor | GoogLeNet | - |
dc.subject.keywordAuthor | Long Short Term Memory | - |
dc.subject.keywordAuthor | Saliency Detection | - |
dc.description.journalRegisteredClass | kci | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
Soongsil University Library 369 Sangdo-Ro, Dongjak-Gu, Seoul, Korea (06978)02-820-0733
COPYRIGHT ⓒ SOONGSIL UNIVERSITY, ALL RIGHTS RESERVED.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.