Learning to Model Relationships for Zero-Shot Video Classification
文献类型:期刊论文
作者 | Gao, Junyu1,2; Zhang, Tianzhu3; Xu, Changsheng1,2 |
刊名 | IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE |
出版日期 | 2021-10-01 |
卷号 | 43期号:10页码:3476-3491 |
ISSN号 | 0162-8828 |
关键词 | Zero-shot video classification graph neural networks zero-shot learning deep attention model |
DOI | 10.1109/TPAMI.2020.2985708 |
通讯作者 | Xu, Changsheng(csxu@nlpr.ia.ac.cn) |
英文摘要 | With the explosive growth of video categories, zero-shot learning (ZSL) in video classification has become a promising research direction in pattern analysis and machine learning. Based on some auxiliary information such as word embeddings and attributes, the key to a robust ZSL method is to transfer the learned knowledge from seen classes to unseen classes, which requires relationship modeling between these concepts (e.g., categories and attributes). However, most existing approaches ignore to model the explicit relationships in an end-to-end manner, resulting in low effectiveness of knowledge transfer. To tackle this problem, we reconsider the video ZSL task as a task-driven message passing process to jointly enjoy several merits including alleviated heterogeneity gap, low domain shift, and robust temporal modeling. Specifically, we propose a prototype-sample GNN (PS-GNN) consisting of a prototype branch and a sample branch to directly and adaptively model all the relationships between category-attribute, category-category, and attribute-attribute. The prototype branch aims to learn robust representations of video categories, which takes as input a set of word-embedding vectors corresponding to the concepts. The sample branch is designed to generate features of a video sample by leveraging its object semantics. With the co-adaption and cooperation between both branches, a unified and robust ZSL framework is achieved. Extensive experiments strongly evidence that PS-GNN obtains favorable performance on five popular video benchmarks consistently. |
资助项目 | National Key Research and Development Program of China[2018AAA0102200] ; National Natural Science Foundation of China[61720106006] ; National Natural Science Foundation of China[61721004] ; National Natural Science Foundation of China[61832002] ; National Natural Science Foundation of China[61532009] ; National Natural Science Foundation of China[U1705262] ; National Natural Science Foundation of China[U1836220] ; National Natural Science Foundation of China[61702511] ; Key Research Program of Frontier Sciences, CAS[QYZDJSSWJSC039] ; Research Program of National Laboratory of Pattern Recognition[Z-2018007] |
WOS研究方向 | Computer Science ; Engineering |
语种 | 英语 |
出版者 | IEEE COMPUTER SOC |
WOS记录号 | WOS:000692232400017 |
资助机构 | National Key Research and Development Program of China ; National Natural Science Foundation of China ; Key Research Program of Frontier Sciences, CAS ; Research Program of National Laboratory of Pattern Recognition |
源URL | [http://ir.ia.ac.cn/handle/173211/46001] |
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.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 3.Univ Sci & Technol China, Hefei 230052, Anhui, Peoples R China |
推荐引用方式 GB/T 7714 | Gao, Junyu,Zhang, Tianzhu,Xu, Changsheng. Learning to Model Relationships for Zero-Shot Video Classification[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2021,43(10):3476-3491. |
APA | Gao, Junyu,Zhang, Tianzhu,&Xu, Changsheng.(2021).Learning to Model Relationships for Zero-Shot Video Classification.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,43(10),3476-3491. |
MLA | Gao, Junyu,et al."Learning to Model Relationships for Zero-Shot Video Classification".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 43.10(2021):3476-3491. |
入库方式: OAI收割
来源:自动化研究所
浏览0
下载0
收藏0
其他版本
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。