-
公开(公告)号：US20190251333A1
公开(公告)日：2019-08-15
申请号：US16392270
申请日：2019-04-23
发明人： Hao Wang , Zhifeng Li , Xing Ji , Fan Jia , Yitong Wang
CPC分类号： G06K9/00228 , G06K9/00 , G06K9/00288 , G06K9/36 , G06K9/46 , G06K9/62 , G06K9/6256 , G06K9/6262 , G06K9/6284 , G06N3/02
摘要： An object detection training method can include receiving a training sample set in a current iteration of an object detection training process over an object detection neural network. The training sample set can include first samples of a first class and second samples of a second class. A first center loss value of each of the first and second samples can be determined. The first center loss value can be a distance between an eigenvector of the respective sample and a center eigenvector of the first or second class which the respective sample belongs to. A second center loss value of the training sample set can be determined according to the first center loss values of the first and second samples. A first target loss value of the current iteration can be determined according to the second center loss value of the training sample set.
-
公开(公告)号：US20210089752A1
公开(公告)日：2021-03-25
申请号：US17109574
申请日：2020-12-02
发明人： Hao WANG , Zhifeng Li , Xing Ji , Fan Jia , Yitong Wang
摘要： An object detection training method can include receiving a training sample set in a current iteration of an object detection training process over an object detection neural network. The training sample set can include first samples of a first class and second samples of a second class. A first center loss value of each of the first and second samples can be determined. The first center loss value can be a distance between a feature vector of the respective sample and a center feature vector of the first or second class which the respective sample belongs to. A second center loss value of the training sample set can be determined according to the first center loss values of the first and second samples. A first target loss value of the current iteration can be determined according to the second center loss value of the training sample set.
-
公开(公告)号：US10929644B2
公开(公告)日：2021-02-23
申请号：US16392270
申请日：2019-04-23
发明人： Hao Wang , Zhifeng Li , Xing Ji , Fan Jia , Yitong Wang
摘要： An object detection training method can include receiving a training sample set in a current iteration of an object detection training process over an object detection neural network. The training sample set can include first samples of a first class and second samples of a second class. A first center loss value of each of the first and second samples can be determined. The first center loss value can be a distance between an eigenvector of the respective sample and a center eigenvector of the first or second class which the respective sample belongs to. A second center loss value of the training sample set can be determined according to the first center loss values of the first and second samples. A first target loss value of the current iteration can be determined according to the second center loss value of the training sample set.
-
公开(公告)号：US11594070B2
公开(公告)日：2023-02-28
申请号：US17109574
申请日：2020-12-02
发明人： Hao Wang , Zhifeng Li , Xing Ji , Fan Jia , Yitong Wang
摘要： An object detection training method can include receiving a training sample set in a current iteration of an object detection training process over an object detection neural network. The training sample set can include first samples of a first class and second samples of a second class. A first center loss value of each of the first and second samples can be determined. The first center loss value can be a distance between a feature vector of the respective sample and a center feature vector of the first or second class which the respective sample belongs to. A second center loss value of the training sample set can be determined according to the first center loss values of the first and second samples. A first target loss value of the current iteration can be determined according to the second center loss value of the training sample set.
-
-
-
搜索结果
4 条记录 (耗时 0.018 秒)
