Question-Guided Erasing-Based Spatiotemporal Attention Learning for Video Question Answering
文献类型:期刊论文
作者 | Liu, Fei2,3; Liu, Jing2,3; Hong, Richang1; Lu, Hanqing2,3 |
刊名 | IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS |
出版日期 | 2021-08-30 |
页码 | 13 |
ISSN号 | 2162-237X |
关键词 | Spatiotemporal phenomena Measurement Visualization Knowledge discovery Task analysis Cognition Semantics Erasing mechanism metric learning spatiotemporal attention video question answering (VideoQA) |
DOI | 10.1109/TNNLS.2021.3105280 |
通讯作者 | Liu, Jing(jliu@nlpr.ia.ac.cn) |
英文摘要 | Spatiotemporal attention learning for video question answering (VideoQA) has always been a challenging task, where existing approaches treat the attention parts and the nonattention parts in isolation. In this work, we propose to enforce the correlation between the attention parts and the nonattention parts as a distance constraint for discriminative spatiotemporal attention learning. Specifically, we first introduce a novel attention-guided erasing mechanism in the traditional spatiotemporal attention to obtain multiple aggregated attention features and nonattention features and then learn to separate the attention and the nonattention features with an appropriate distance. The distance constraint is enforced by a metric learning loss, without increasing the inference complexity. In this way, the model can learn to produce more discriminative spatiotemporal attention distribution on videos, thus enabling more accurate question answering. In order to incorporate the multiscale spatiotemporal information that is beneficial for video understanding, we additionally develop a pyramid variant on basis of the proposed approach. Comprehensive ablation experiments are conducted to validate the effectiveness of our approach, and state-of-the-art performance is achieved on several widely used datasets for VideoQA. |
WOS关键词 | IMAGE SIMILARITY |
资助项目 | National Key Research and Development Program of China[2020AAA0106400] ; National Natural Science Foundation of China[61922086] ; National Natural Science Foundation of China[61872366] |
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:000733489300001 |
资助机构 | National Key Research and Development Program of China ; National Natural Science Foundation of China |
源URL | [http://ir.ia.ac.cn/handle/173211/47028] |
专题 | 自动化研究所_模式识别国家重点实验室_图像与视频分析团队 |
通讯作者 | Liu, Jing |
作者单位 | 1.Hefei Univ Technol, Sch Comp & Informat, Hefei 230000, Peoples R China 2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China |
推荐引用方式 GB/T 7714 | Liu, Fei,Liu, Jing,Hong, Richang,et al. Question-Guided Erasing-Based Spatiotemporal Attention Learning for Video Question Answering[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2021:13. |
APA | Liu, Fei,Liu, Jing,Hong, Richang,&Lu, Hanqing.(2021).Question-Guided Erasing-Based Spatiotemporal Attention Learning for Video Question Answering.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,13. |
MLA | Liu, Fei,et al."Question-Guided Erasing-Based Spatiotemporal Attention Learning for Video Question Answering".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2021):13. |
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。