专利内容由知识产权出版社提供
专利名称：LEARNING METHOD AND LEARNING DEVICE
FOR IMPROVING PERFORMANCE OF CNNBY USING FEATURE UPSAMPLING
NETWORKS, AND TESTING METHOD ANDTESTING DEVICE USING THE SAME
发明人：KIM, Yongjoong,NAM, Woonhyun,BOO,
Sukhoon,SUNG, Myungchul,YEO,Donghun,RYU, Wooju,JANG,Taewoong,JEONG, Kyungjoong,JE,Hongmo,CHO, Hojin
申请号：EP18192817.7申请日：20180905公开号：EP3467720A1公开日：20190410
专利附图：
摘要：A learning method for improving performance of a CNN by using Feature Up-sampling Networks is disclosed. The learning method includes steps of: (a) allowing thedown-sampling block to acquire a down-sampling image; (b) allowing each of a (1-1)-th toa (1-k)-th filter blocks to respectively acquire each of a (1-1)-th to a (1-k)-th feature maps;(c) allowing a specific up-sampling block to (i) receive a particular feature map from itscorresponding filter block, and (ii) receive another specific feature map from its previousup-sampling block, and then rescale a size of the specific feature map to be identical tothat of the particular feature map and (iii) apply certain operations to the particularfeature map and the rescaled specific feature map to generate a feature map of thespecific up-sampling block; and (d)(i) allowing an application block to acquire anapplication-specific output and (ii) performing a first backpropagation process.
申请人：StradVision, Inc.
地址：4427, Complex 4, 67 Cheongam-ro, Nam-gu Pohang-si Gyeongsangbuk-do37673 KR
国籍：KR
代理机构：Klunker IP Patentanwälte PartG mbB
更多信息请下载全文后查看