BriefGPT.xyz
Mar, 2023
利用语义属性进行归纳零样本学习
Exploiting Semantic Attributes for Transductive Zero-Shot Learning
HTML
PDF
Zhengbo Wang, Jian Liang, Zilei Wang, Tieniu Tan
TL;DR
本文提出一种新的传输零样本学习方法,该方法使用生成对抗网络从未见过的特征提取出语义属性,并将其融合到产生模型中,从而捕获未见类别内的细微差异,合成更具辨别能力的特征。在五个标准基准测试中,本方法取得了零样本学习的最新成果。
Abstract
zero-shot learning
(ZSL) aims to recognize unseen classes by generalizing the relation between visual features and
semantic attributes
learned from the seen classes. A recent paradigm called transductive
→