摘要
阐述针对深度学习网络中深层语义特性的分析,提出不完整多模态数据融合方法,借助多模态的高层语义,构建一个统一的深度学习模型,既可以融合模态私有深度网络,又可以共享模态特征。
This paper the analysis of deep semantic characteristics in deep learning networks,proposes an incomplete multimodal data fusion method, and utilizes the high-level semantics of multimodality to construct a unified deep learning model, which can integrate modal private deep networks and share modal features.
作者
夏之罡
陆文彪
叶盛
王骏永
XIA Zhigang;LU Wenbiao;YE Sheng;WANG Junyong(Lishui Qiaoqi Electric Power Industry Group Co.,Ltd.,Zhejiang 323000,China)
出处
《电子技术（上海）》
2024年第3期246-247,共2页
Electronic Technology
关键词
深度学习网络
多模态数据
智能算法
deep learning networks
multimodal data
intelligent algorithms
