R. Vrskova, R. Hudec, P. Sykora, P. Kamencay, M. Radilova
{"title":"Education of Video Classification Based by Neural Networks","authors":"R. Vrskova, R. Hudec, P. Sykora, P. Kamencay, M. Radilova","doi":"10.1109/ICETA51985.2020.9379190","DOIUrl":null,"url":null,"abstract":"In this paper an artificial neural network for video classification were presented. Artificial neural networks were essential part of the video classification. As they are more and more used for applications doing of video classification, it is desirable to introduce them in the educational process. After introduction to the topic, basic theory about architecture neural networks for video classification continues. Firstly, the 3D Convolution Neural network (3DCNN) using UCF50-action recognition database was applied. Next the Convolutional Long Short-Term Memory (ConvLSTM) on the same dataset was used. Finally, these neural networks using confusion matrix were compared. The all experimental results using UCF50-Datatset were performed. The achieved experimental results demonstrate the effectiveness of neural networks (3D CNN and ConvLSTM) in educational process.","PeriodicalId":149716,"journal":{"name":"2020 18th International Conference on Emerging eLearning Technologies and Applications (ICETA)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 18th International Conference on Emerging eLearning Technologies and Applications (ICETA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICETA51985.2020.9379190","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
In this paper an artificial neural network for video classification were presented. Artificial neural networks were essential part of the video classification. As they are more and more used for applications doing of video classification, it is desirable to introduce them in the educational process. After introduction to the topic, basic theory about architecture neural networks for video classification continues. Firstly, the 3D Convolution Neural network (3DCNN) using UCF50-action recognition database was applied. Next the Convolutional Long Short-Term Memory (ConvLSTM) on the same dataset was used. Finally, these neural networks using confusion matrix were compared. The all experimental results using UCF50-Datatset were performed. The achieved experimental results demonstrate the effectiveness of neural networks (3D CNN and ConvLSTM) in educational process.