CASIA OpenIR  > 模式识别国家重点实验室  > 多媒体计算与图形学
Deep Structured Event Modeling for User-Generated Photos
Yang, Xiaoshan1,2; Zhang, Tianzhu1,2; Xu, Changsheng1,2
Source PublicationIEEE TRANSACTIONS ON MULTIMEDIA
2018
Volume20Issue:8Pages:2100-2113
AbstractVision based event analysis plays a very critical role in automatically organizing user generated photos. It is difficult because of the following challenges. (1) Intra-class variation. Photos uploaded by users are sparsely sampled visual appearances of event over time. Thus, each photo may only capture a single object or scene of a specific complex event. (2) Inter-class confusion. Photos related to different events may contain similar objects or scenes. (3) For abnormal event, it has sample scarcity issue and only few samples can be used for event pattern learning. In this paper, by considering the photo taken time, we propose a novel structured event modeling (SEM) framework for event analysis by exploiting temporal information of visual features and event classes in photo sequence. More specifically, the temporal event patterns of the photo sequence and the relationship of different photos are learned jointly through deep neural networks (CNNs and RNNs) and conditional random field (CRF). We evaluate the proposed structured event modeling framework in two applications including the mutliclass event recognition and abnormal event discovery in photo sequence. The extensive experimental results on a public event recognition dataset and a collected abnormal event dataset demonstrate the effectiveness of the proposed method.
KeywordEvent Analysis Unusual Event Detection Deep Learning
DOI10.1109/TMM.2017.2788210
Indexed BySCI
Language英语
Citation statistics
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/19702
Collection模式识别国家重点实验室_多媒体计算与图形学
Affiliation1.National Lab of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences
2.University of Chinese Academy of Sciences
First Author AffilicationInstitute of Automation, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
Yang, Xiaoshan,Zhang, Tianzhu,Xu, Changsheng. Deep Structured Event Modeling for User-Generated Photos[J]. IEEE TRANSACTIONS ON MULTIMEDIA,2018,20(8):2100-2113.
APA Yang, Xiaoshan,Zhang, Tianzhu,&Xu, Changsheng.(2018).Deep Structured Event Modeling for User-Generated Photos.IEEE TRANSACTIONS ON MULTIMEDIA,20(8),2100-2113.
MLA Yang, Xiaoshan,et al."Deep Structured Event Modeling for User-Generated Photos".IEEE TRANSACTIONS ON MULTIMEDIA 20.8(2018):2100-2113.
Files in This Item: Download All
File Name/Size DocType Version Access License
Deep-Structured Even(1164KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Yang, Xiaoshan]'s Articles
[Zhang, Tianzhu]'s Articles
[Xu, Changsheng]'s Articles
Baidu academic
Similar articles in Baidu academic
[Yang, Xiaoshan]'s Articles
[Zhang, Tianzhu]'s Articles
[Xu, Changsheng]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Yang, Xiaoshan]'s Articles
[Zhang, Tianzhu]'s Articles
[Xu, Changsheng]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Deep-Structured Event Modeling.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.