利用ResNet实现生产线上工人装配动作的自动视频标注

Myke D. M. Valadão, Diego A. Amoedo, Gustavo M. Torres, E. V. C. U. Mattos, Antônio M. C. Pereira, Matheus S. Uchôa, Lucas M. Torres, Victor L. G. Cavalcante, José E. B. S. Linhares, M. O. Silva, Agemilson P. Silva, Caio F. S. Cruz, Rômulo Fabrício, Ruan J. S. Belém, Thiago B. Bezerra, W. S. S. Júnior, Celso B. Carvalho
{"title":"利用ResNet实现生产线上工人装配动作的自动视频标注","authors":"Myke D. M. Valadão, Diego A. Amoedo, Gustavo M. Torres, E. V. C. U. Mattos, Antônio M. C. Pereira, Matheus S. Uchôa, Lucas M. Torres, Victor L. G. Cavalcante, José E. B. S. Linhares, M. O. Silva, Agemilson P. Silva, Caio F. S. Cruz, Rômulo Fabrício, Ruan J. S. Belém, Thiago B. Bezerra, W. S. S. Júnior, Celso B. Carvalho","doi":"10.1109/ICCE-Taiwan55306.2022.9869008","DOIUrl":null,"url":null,"abstract":"In this work, conducted by two partners, called UFAM/CETELI and, Envision (TPV Group), we present a method of automatic labeling of frames of worker's actions in factory environments using a model generated by a residual neural network. With this approach we used some manually labeled frames to training a model that provide the label of 4 classes of actions. We achieve accuracy rate over 96%, which give reliability to a supervised training of 3D dataset of actions.","PeriodicalId":164671,"journal":{"name":"2022 IEEE International Conference on Consumer Electronics - Taiwan","volume":"28 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Automatic Video Labeling with Assembly Actions of Workers on a Production Line Using ResNet\",\"authors\":\"Myke D. M. Valadão, Diego A. Amoedo, Gustavo M. Torres, E. V. C. U. Mattos, Antônio M. C. Pereira, Matheus S. Uchôa, Lucas M. Torres, Victor L. G. Cavalcante, José E. B. S. Linhares, M. O. Silva, Agemilson P. Silva, Caio F. S. Cruz, Rômulo Fabrício, Ruan J. S. Belém, Thiago B. Bezerra, W. S. S. Júnior, Celso B. Carvalho\",\"doi\":\"10.1109/ICCE-Taiwan55306.2022.9869008\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this work, conducted by two partners, called UFAM/CETELI and, Envision (TPV Group), we present a method of automatic labeling of frames of worker's actions in factory environments using a model generated by a residual neural network. With this approach we used some manually labeled frames to training a model that provide the label of 4 classes of actions. We achieve accuracy rate over 96%, which give reliability to a supervised training of 3D dataset of actions.\",\"PeriodicalId\":164671,\"journal\":{\"name\":\"2022 IEEE International Conference on Consumer Electronics - Taiwan\",\"volume\":\"28 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-07-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 IEEE International Conference on Consumer Electronics - Taiwan\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICCE-Taiwan55306.2022.9869008\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE International Conference on Consumer Electronics - Taiwan","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCE-Taiwan55306.2022.9869008","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

在这项由两个合作伙伴UFAM/CETELI和Envision(冠捷集团)进行的工作中,我们提出了一种使用残量神经网络生成的模型自动标记工厂环境中工人动作框架的方法。通过这种方法,我们使用一些手动标记的帧来训练一个模型,该模型提供了4类动作的标签。我们实现了96%以上的准确率,这为三维动作数据集的监督训练提供了可靠性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Automatic Video Labeling with Assembly Actions of Workers on a Production Line Using ResNet
In this work, conducted by two partners, called UFAM/CETELI and, Envision (TPV Group), we present a method of automatic labeling of frames of worker's actions in factory environments using a model generated by a residual neural network. With this approach we used some manually labeled frames to training a model that provide the label of 4 classes of actions. We achieve accuracy rate over 96%, which give reliability to a supervised training of 3D dataset of actions.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Dynamic Thermal-Predicted Workload Movement with Three-Dimensional DRAM-RRAM Hybrid Memories for Convolutional Neural Network Applications Performance Evaluation of Fault-Tolerant Routing Methods Using Parallel Programs Down-Sampling Dark Channel Prior of Airlight Estimation for Low Complexity Image Dehazing Chip Design Image Confusion Applied to Industrial Defect Detection System On Multimodal Semantic Consistency Detection of News Articles with Image Caption Pairs
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1