Dual aggregated feature pyramid network for multi label classification
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Yun, Dongjoo | - |
dc.contributor.author | Ryu, Jongbin | - |
dc.contributor.author | Lim, Jongwoo | - |
dc.date.accessioned | 2022-07-06T22:39:30Z | - |
dc.date.available | 2022-07-06T22:39:30Z | - |
dc.date.created | 2021-05-11 | - |
dc.date.issued | 2021-04 | - |
dc.identifier.issn | 0167-8655 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/142135 | - |
dc.description.abstract | While many deep convolutional neural networks show promising performance in various classification tasks, multiple objects appearing in very different sizes, shapes, and appearances cause difficulty in multilabel classification using conventional neural networks. In this paper, we introduce a dual aggregated network on pyramidal convolutional features for multi-label classification. The proposed method includes both featureand classifier-level aggregation to learn discriminant multi-scale information of various target objects in the image. First, the feature-level aggregation collects the convolutional activation maps from the multi-scale pyramid network, and then it densely pools them to take localized features of each object. We elaborately design the feature aggregation method so that the responses from the objects with different sizes, aspect ratios, and shapes are properly reflected the aggregated activation map. Unlike conventional methods, this process does not require the region proposal step, which reduces the computational burden significantly. Second, we introduce the classifier level aggregation algorithm for integrating the multi-object classifier modules. To maximize the discrimination power of each class, we train one-vs-all classifiers for individual classes using the class-wise loss function. For each test image, the scores from the class-wise classifiers are aggregated to get the final multi-label classification result. By combining the above featureand classifier-level aggregation methods, our network can be trained in an end-to-end fashion, which is not possible for the conventional multi-label classification algorithms using region proposals. Extensive evaluations on PASCAL VOC 2007 and PASCAL VOC 2012 demonstrate that the proposed algorithm outperforms the state-of-the-art methods. | - |
dc.language | 영어 | - |
dc.language.iso | en | - |
dc.publisher | Elsevier B.V. | - |
dc.title | Dual aggregated feature pyramid network for multi label classification | - |
dc.type | Article | - |
dc.contributor.affiliatedAuthor | Lim, Jongwoo | - |
dc.identifier.doi | 10.1016/j.patrec.2021.01.013 | - |
dc.identifier.scopusid | 2-s2.0-85100439913 | - |
dc.identifier.wosid | 000624287500011 | - |
dc.identifier.bibliographicCitation | Pattern Recognition Letters, v.144, pp.75 - 81 | - |
dc.relation.isPartOf | Pattern Recognition Letters | - |
dc.citation.title | Pattern Recognition Letters | - |
dc.citation.volume | 144 | - |
dc.citation.startPage | 75 | - |
dc.citation.endPage | 81 | - |
dc.type.rims | ART | - |
dc.type.docType | Article | - |
dc.description.journalClass | 1 | - |
dc.description.isOpenAccess | N | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Artificial Intelligence | - |
dc.subject.keywordPlus | Aspect ratio | - |
dc.subject.keywordPlus | Chemical activation | - |
dc.subject.keywordPlus | Convolution | - |
dc.subject.keywordPlus | Convolutional neural networks | - |
dc.subject.keywordPlus | Deep neural networks | - |
dc.subject.keywordPlus | Aggregation algorithms | - |
dc.subject.keywordPlus | Classification tasks | - |
dc.subject.keywordPlus | Computational burden | - |
dc.subject.keywordPlus | Conventional methods | - |
dc.subject.keywordPlus | Multi label classification | - |
dc.subject.keywordPlus | Multi-scale informations | - |
dc.subject.keywordPlus | Multi-Scale pyramids | - |
dc.subject.keywordPlus | State-of-the-art methods | - |
dc.subject.keywordPlus | Classification (of information) | - |
dc.subject.keywordAuthor | Aggregation | - |
dc.subject.keywordAuthor | Deep learning | - |
dc.subject.keywordAuthor | Multi-label classification | - |
dc.identifier.url | https://www.sciencedirect.com/science/article/pii/S016786552100026X?via%3Dihub | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
222, Wangsimni-ro, Seongdong-gu, Seoul, 04763, Korea+82-2-2220-1365
COPYRIGHT © 2021 HANYANG UNIVERSITY.
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.