Full metadata record
DC Field | Value | Language |
---|---|---|
dc.citation.number | 4 | - |
dc.citation.startPage | 1464 | - |
dc.citation.title | SENSORS | - |
dc.citation.volume | 22 | - |
dc.contributor.author | Saqlain, Muhammad | - |
dc.contributor.author | Kim, Donguk | - |
dc.contributor.author | Cha, Junuk | - |
dc.contributor.author | Lee, Changhwa | - |
dc.contributor.author | Lee, Seongyeong | - |
dc.contributor.author | Baek, Seungryul | - |
dc.date.accessioned | 2023-12-21T14:37:42Z | - |
dc.date.available | 2023-12-21T14:37:42Z | - |
dc.date.created | 2022-04-11 | - |
dc.date.issued | 2022-02 | - |
dc.description.abstract | Group activity recognition is a prime research topic in video understanding and has many practical applications, such as crowd behavior monitoring, video surveillance, etc. To understand the multi-person/group action, the model should not only identify the individual person's action in the context but also describe their collective activity. A lot of previous works adopt skeleton-based approaches with graph convolutional networks for group activity recognition. However, these approaches are subject to limitation in scalability, robustness, and interoperability. In this paper, we propose 3DMesh-GAR, a novel approach to 3D human body Mesh-based Group Activity Recognition, which relies on a body center heatmap, camera map, and mesh parameter map instead of the complex and noisy 3D skeleton of each person of the input frames. We adopt a 3D mesh creation method, which is conceptually simple, single-stage, and bounding box free, and is able to handle highly occluded and multi-person scenes without any additional computational cost. We implement 3DMesh-GAR on a standard group activity dataset: the Collective Activity Dataset, and achieve state-of-the-art performance for group activity recognition. | - |
dc.identifier.bibliographicCitation | SENSORS, v.22, no.4, pp.1464 | - |
dc.identifier.doi | 10.3390/s22041464 | - |
dc.identifier.issn | 1424-8220 | - |
dc.identifier.scopusid | 2-s2.0-85124486058 | - |
dc.identifier.uri | https://scholarworks.unist.ac.kr/handle/201301/58151 | - |
dc.identifier.url | https://www.mdpi.com/1424-8220/22/4/1464 | - |
dc.identifier.wosid | 000771878700001 | - |
dc.language | 영어 | - |
dc.publisher | MDPI | - |
dc.title | 3DMesh-GAR: 3D Human Body Mesh-Based Method for Group Activity Recognition | - |
dc.type | Article | - |
dc.description.isOpenAccess | TRUE | - |
dc.relation.journalWebOfScienceCategory | Chemistry, Analytical; Engineering, Electrical & Electronic; Instruments & Instrumentation | - |
dc.relation.journalResearchArea | Chemistry; Engineering; Instruments & Instrumentation | - |
dc.type.docType | Article | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.subject.keywordAuthor | 3D human activity recognition | - |
dc.subject.keywordAuthor | human body mesh estimation | - |
dc.subject.keywordAuthor | feature extraction | - |
dc.subject.keywordAuthor | deep learning | - |
dc.subject.keywordAuthor | video understanding | - |
Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Tel : 052-217-1404 / Email : scholarworks@unist.ac.kr
Copyright (c) 2023 by UNIST LIBRARY. All rights reserved.
ScholarWorks@UNIST was established as an OAK Project for the National Library of Korea.