专利名称：Supervid domain adaptation
发明人：Martina Marek,Stefan H尚小云
interstoisr,Olivier
Pauly,Hauke Heibel,Martin Bokeloh
申请号：US17096691
申请日：20201112
公开号：US11170581B1
公开日：
20211109
专利内容由知识产权出版社提供
专利附图：
摘要：Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a feature extraction neural network to generate domain-invariant feature reprentations from domain-varyin在后面用英语怎么说
g input 上海抗疫
images. In one
aspect, the method includes obtaining a training datat comprising a first t of target domain images and a cond t of 亚麻色
real domai哈尔滨工业大学录取分数线
n images that each have pixel-wi leve逢开头的成语
l alignment with a corresponding target do萧红介绍
main image, and training the feature extraction n中国火车发展史
eural network on the training datat bad on optimizing an objective function that includes a term that depends on a simil顶针怎么用
arity between respective featur论文致谢词
e reprentations generated by the network for a pair of target and source domain images.
申请人：Intrinsic Innovation LLC
地址：Mountain View CA US
国籍：US
代理机构：Fish & Richardson P.C.
更多信息请下载全文后查看
