Action Recognition Using Multi-stream 2D CNN with Deep Learning-Based Temporal Modality
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kang, K. | - |
dc.contributor.author | Park, S. | - |
dc.contributor.author | Park, H. | - |
dc.contributor.author | Kang, D. | - |
dc.contributor.author | Paik, Joon Ki | - |
dc.date.accessioned | 2023-05-17T01:40:57Z | - |
dc.date.available | 2023-05-17T01:40:57Z | - |
dc.date.issued | 2023-01 | - |
dc.identifier.issn | 0747-668X | - |
dc.identifier.uri | https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/66537 | - |
dc.description.abstract | Video action recognition requires accurate analysis of motion information along with spatial information of an object. In other words, it is necessary to learn both temporal and spatial information. In many deep learning-based action recognition methods, temporal and spatial information are extracted by a multi-stream network, where the temporal stream network analyzes the motion information using mathematical operations. In this paper, we present an action recognition method using a multi-stream network with a deep learning-based temporal relation module, which extracts motion information for the entire video in the temporal network path. The proposed method significantly increases the accuracy of action recognition using attached modules in front of the 2D CNN and late fusion with another network path. Owing to the proposed temporal stream network without additional mathematical operations, we could greatly reduces the amount of computation. As a result, the proposed method is suitable for a wide range of real-time visual action recognition tasks. © 2023 IEEE. | - |
dc.language | 영어 | - |
dc.language.iso | ENG | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.title | Action Recognition Using Multi-stream 2D CNN with Deep Learning-Based Temporal Modality | - |
dc.type | Article | - |
dc.identifier.doi | 10.1109/ICCE56470.2023.10043568 | - |
dc.identifier.bibliographicCitation | Digest of Technical Papers - IEEE International Conference on Consumer Electronics, v.2023-January | - |
dc.description.isOpenAccess | N | - |
dc.identifier.wosid | 000978390700191 | - |
dc.identifier.scopusid | 2-s2.0-85149112306 | - |
dc.citation.title | Digest of Technical Papers - IEEE International Conference on Consumer Electronics | - |
dc.citation.volume | 2023-January | - |
dc.type.docType | Proceedings Paper | - |
dc.subject.keywordAuthor | Action recognition | - |
dc.subject.keywordAuthor | Temporal modality | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalResearchArea | Engineering | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Interdisciplinary Applications | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.description.journalRegisteredClass | scopus | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
84, Heukseok-ro, Dongjak-gu, Seoul, Republic of Korea (06974)02-820-6194
COPYRIGHT 2019 Chung-Ang 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.