{"title":"基于改进中心网的人脸特征检测方法","authors":"Zhou Sheng-an","doi":"10.1109/ICSCDE54196.2021.00048","DOIUrl":null,"url":null,"abstract":"Aiming at the problems of CenterNet detection model with many parameters, long convergence time and slow detection speed, this paper proposed an real time detection method. In this method, mobileNet is used as the back-end to extract the features of the scene before the image, and then combined with the lightweight CentrerNet detection head to detect. Based on CelebA and MSCOCO datasets, the results show thatOur Method model has the best balance in trainning time, detection speed and accuracy, and better application value compared with the improved reference model of CenterNet_ Shuffler and CenterNet_ ResNet18.","PeriodicalId":208108,"journal":{"name":"2021 International Conference of Social Computing and Digital Economy (ICSCDE)","volume":"126 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Facial features detection method based on improved centernet\",\"authors\":\"Zhou Sheng-an\",\"doi\":\"10.1109/ICSCDE54196.2021.00048\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Aiming at the problems of CenterNet detection model with many parameters, long convergence time and slow detection speed, this paper proposed an real time detection method. In this method, mobileNet is used as the back-end to extract the features of the scene before the image, and then combined with the lightweight CentrerNet detection head to detect. Based on CelebA and MSCOCO datasets, the results show thatOur Method model has the best balance in trainning time, detection speed and accuracy, and better application value compared with the improved reference model of CenterNet_ Shuffler and CenterNet_ ResNet18.\",\"PeriodicalId\":208108,\"journal\":{\"name\":\"2021 International Conference of Social Computing and Digital Economy (ICSCDE)\",\"volume\":\"126 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-08-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 International Conference of Social Computing and Digital Economy (ICSCDE)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICSCDE54196.2021.00048\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 International Conference of Social Computing and Digital Economy (ICSCDE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICSCDE54196.2021.00048","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Facial features detection method based on improved centernet
Aiming at the problems of CenterNet detection model with many parameters, long convergence time and slow detection speed, this paper proposed an real time detection method. In this method, mobileNet is used as the back-end to extract the features of the scene before the image, and then combined with the lightweight CentrerNet detection head to detect. Based on CelebA and MSCOCO datasets, the results show thatOur Method model has the best balance in trainning time, detection speed and accuracy, and better application value compared with the improved reference model of CenterNet_ Shuffler and CenterNet_ ResNet18.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
