VMAN: A Virtual Mainstay Alignment Network for Transductive Zero-Shot Learning | |
Xie, Guo-Sen5,6; Zhang, Xu-Yao1; Yao, Yazhou5; Zhang, Zheng2,3; Zhao, Fang4; Shao, Ling4 | |
刊名 | IEEE TRANSACTIONS ON IMAGE PROCESSING |
2021 | |
卷号 | 30页码:4316-4329 |
关键词 | Semantics Training Task analysis Image reconstruction Whales Manifolds Generative adversarial networks Zero-shot learning virtual sample generation transductive |
ISSN号 | 1057-7149 |
DOI | 10.1109/TIP.2021.3070231 |
通讯作者 | Yao, Yazhou(yazhou.yao@njust.edu.cn) |
英文摘要 | Transductive zero-shot learning (TZSL) extends conventional ZSL by leveraging (unlabeled) unseen images for model training. A typical method for ZSL involves learning embedding weights from the feature space to the semantic space. However, the learned weights in most existing methods are dominated by seen images, and can thus not be adapted to unseen images very well. In this paper, to align the (embedding) weights for better knowledge transfer between seen/unseen classes, we propose the virtual mainstay alignment network (VMAN), which is tailored for the transductive ZSL task. Specifically, VMAN is casted as a tied encoder-decoder net, thus only one linear mapping weights need to be learned. To explicitly learn the weights in VMAN, for the first time in ZSL, we propose to generate virtual mainstay (VM) samples for each seen class, which serve as new training data and can prevent the weights from being shifted to seen images, to some extent. Moreover, a weighted reconstruction scheme is proposed and incorporated into the model training phase, in both the semantic/feature spaces. In this way, the manifold relationships of the VM samples are well preserved. To further align the weights to adapt to more unseen images, a novel instance-category matching regularization is proposed for model re-training. VMAN is thus modeled as a nested minimization problem and is solved by a Taylor approximate optimization paradigm. In comprehensive evaluations on four benchmark datasets, VMAN achieves superior performances under the (Generalized) TZSL setting. |
资助项目 | National Natural Science Foundation of China[61702163] ; National Natural Science Foundation of China[61976116] ; National Natural Science Foundation of China[62002085] ; Fundamental Research Funds for the Central Universities[30920021135] |
WOS研究方向 | Computer Science ; Engineering |
语种 | 英语 |
出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
WOS记录号 | WOS:000641960800002 |
资助机构 | National Natural Science Foundation of China ; Fundamental Research Funds for the Central Universities |
内容类型 | 期刊论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/44494] |
专题 | 自动化研究所_模式识别国家重点实验室_模式分析与学习团队 |
通讯作者 | Yao, Yazhou |
作者单位 | 1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Harbin Inst Technol, Shenzhen Key Lab Visual Object Detect & Recognit, Shenzhen 518055, Peoples R China 3.Peng Cheng Lab, Shenzhen 518055, Peoples R China 4.Incept Inst Artificial Intelligence, Abu Dhabi, U Arab Emirates 5.Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Nanjing 210094, Peoples R China 6.Mohamed Bin Zayed Univ Artificial Intelligence, Machine Learning Dept, Abu Dhabi, U Arab Emirates |
推荐引用方式 GB/T 7714 | Xie, Guo-Sen,Zhang, Xu-Yao,Yao, Yazhou,et al. VMAN: A Virtual Mainstay Alignment Network for Transductive Zero-Shot Learning[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2021,30:4316-4329. |
APA | Xie, Guo-Sen,Zhang, Xu-Yao,Yao, Yazhou,Zhang, Zheng,Zhao, Fang,&Shao, Ling.(2021).VMAN: A Virtual Mainstay Alignment Network for Transductive Zero-Shot Learning.IEEE TRANSACTIONS ON IMAGE PROCESSING,30,4316-4329. |
MLA | Xie, Guo-Sen,et al."VMAN: A Virtual Mainstay Alignment Network for Transductive Zero-Shot Learning".IEEE TRANSACTIONS ON IMAGE PROCESSING 30(2021):4316-4329. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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。
修改评论