赵亮

个人信息Personal Information

副教授

博士生导师

硕士生导师

主要任职:无

性别:男

毕业院校:大连理工大学

学位:博士

所在单位:软件学院、国际信息与软件学院

学科:软件工程

办公地点:软件学院综合楼417

联系方式:liangzhao@dlut.edu.cn

扫描关注

论文成果

当前位置: 中文主页 >> 科学研究 >> 论文成果

Deep Semantic Mapping for Heterogeneous Multimedia Transfer Learning Using Co-Occurrence Data

点击次数:

论文类型:期刊论文

第一作者:Zhao, Liang

通讯作者:Zhao, L (reprint author), Dalian Univ Technol, Sch Software Technol, Key Lab Ubiquitous Network & Serv Software Liaoni, Dalian 116024, Liaoning, Peoples R China.

合写作者:Chen, Zhikui,Yang, Laurence T.,Deen, M. Jamal,Wang, Z. Jane

发表时间:2019-02-01

发表刊物:ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS

收录刊物:SCIE、EI

卷号:15

期号:1

ISSN号:1551-6857

关键字:Deep semantic mapping; heterogeneous multimedia; transfer learning; deep neural networks; canonical correlation analysis

摘要:Transfer learning, which focuses on finding a favorable representation for instances of different domains based on auxiliary data, can mitigate the divergence between domains through knowledge transfer. Recently, increasing efforts on transfer learning have employed deep neural networks (DNN) to learn more robust and higher level feature representations to better tackle cross-media disparities. However, only a few articles consider the correction and semantic matching between multi-layer heterogeneous domain networks. In this article, we propose a deep semantic mapping model for heterogeneous multimedia transfer learning (DHTL) using co-occurrence data. More specifically, we integrate the DNN with canonical correlation analysis (CCA) to derive a deep correlation subspace as the joint semantic representation for associating data across different domains. In the proposed DHTL, a multi-layer correlation matching network across domains is constructed, in which the CCA is combined to bridge each pair of domain-specific hidden layers. To train the network, a joint objective function is defined and the optimization processes are presented. When the deep semantic representation is achieved, the shared features of the source domain are transferred for task learning in the target domain. Extensive experiments for three multimedia recognition applications demonstrate that the proposed DHTL can effectively find deep semantic representations for heterogeneous domains, and it is superior to the several existing state-of-the-art methods for deep transfer learning.