The Model May Fit You: User-Generalized Cross-Modal Retrieval
Ma, Xinhong2,3; Yang, Xiaoshan1,2,3; Gao, Junyu2,3; Xu, Changsheng1,2,3
刊名IEEE TRANSACTIONS ON MULTIMEDIA
2021-06-25
卷号24页码:2998-3012
关键词cross-modal retrieval domain generalization meta-learning
ISSN号1520-9210
英文摘要

In real-world applications, a cross-model retrieval model trained on multimodal instances without considering differences in data distributions among users, termed as user domain shift, usually cannot generalize well to unknown user domains. In this paper, we define a new task of user-generalized cross-modal retrieval, and propose a novel Meta-Learning Multimodal User Generalization (MLMUG) method to solve it. MLMUG simulates the user domain shift with meta-optimization, which aims to embed multimodal data effectively and generalize the cross-modal retrieval model to any unknown user domains. We design a cross-modal embedding network with a learnable meta covariant attention module to encode transferable knowledge among different user domains. A user-adaptive meta-optimization scheme is proposed to adaptively aggregate gradients and meta-gradients for fast and stable meta-optimization. We build two benchmarks for user-generalized cross-modal retrieval evaluation. Experiments on the proposed benchmarks validate the generalization of our method compared with several state-of-the-art methods.

语种英语
内容类型期刊论文
源URL[http://ir.ia.ac.cn/handle/173211/48795]  
专题自动化研究所_模式识别国家重点实验室_多媒体计算与图形学团队
通讯作者Xu, Changsheng
作者单位1.Peng Cheng Laboratory, Shenzhen, China
2.School of Artificial Intelligence, University of Chinese Academy of Sciences
3.National Lab of Pattern Recognition (NLPR), Institute of Automation, Chinese Academy of Sciences (CASIA)
推荐引用方式
GB/T 7714
Ma, Xinhong,Yang, Xiaoshan,Gao, Junyu,et al. The Model May Fit You: User-Generalized Cross-Modal Retrieval[J]. IEEE TRANSACTIONS ON MULTIMEDIA,2021,24:2998-3012.
APA Ma, Xinhong,Yang, Xiaoshan,Gao, Junyu,&Xu, Changsheng.(2021).The Model May Fit You: User-Generalized Cross-Modal Retrieval.IEEE TRANSACTIONS ON MULTIMEDIA,24,2998-3012.
MLA Ma, Xinhong,et al."The Model May Fit You: User-Generalized Cross-Modal Retrieval".IEEE TRANSACTIONS ON MULTIMEDIA 24(2021):2998-3012.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace