CASIA OpenIR  > 模式识别国家重点实验室  > 视频内容安全
Horror Image Recognition Based on Context-Aware Multi-Instance Learning
Li, Bing1; Xiong, Weihua1; Wu, Ou1; Hu, Weiming1; Maybank, Stephen2; Yan, Shuicheng3; Bing Li
2015-12-01
发表期刊IEEE TRANSACTIONS ON IMAGE PROCESSING
卷号24期号:12页码:5193-5205
文章类型Article
摘要Horror content sharing on the Web is a growing phenomenon that can interfere with our daily life and affect the mental health of those involved. As an important form of expression, horror images have their own characteristics that can evoke extreme emotions. In this paper, we present a novel context-aware multi-instance learning (CMIL) algorithm for horror image recognition. The CMIL algorithm identifies horror images and picks out the regions that cause the sensation of horror in these horror images. It obtains contextual cues among adjacent regions in an image using a random walk on a contextual graph. Borrowing the strength of the fuzzy support vector machine (FSVM), we define a heuristic optimization procedure based on the FSVM to search for the optimal classifier for the CMIL. To improve the initialization of the CMIL, we propose a novel visual saliency model based on the tensor analysis. The average saliency value of each segmented region is set as its initial fuzzy membership in the CMIL. The advantage of the tensor-based visual saliency model is that it not only adaptively selects features, but also dynamically determines fusion weights for saliency value combination from different feature subspaces. The effectiveness of the proposed CMIL model is demonstrated by its use in horror image recognition on two large-scale image sets collected from the Internet.
关键词Horror Image Recognition Context-aware Multi-instance Learning Visual Saliency
WOS标题词Science & Technology ; Technology
DOI10.1109/TIP.2015.2479400
关键词[WOS]SALIENT REGION DETECTION ; VISUAL-ATTENTION ; FEAR-ACQUISITION ; COLOR ; TEXTURE ; SEGMENTATION ; RETRIEVAL ; EMOTION ; MODEL ; INFORMATION
收录类别SCI ; SSCi
语种英语
WOS研究方向Computer Science ; Engineering
WOS类目Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号WOS:000362488900007
引用统计
被引频次:6[WOS]   [WOS记录]     [WOS相关记录]
文献类型期刊论文
条目标识符http://ir.ia.ac.cn/handle/173211/10029
专题模式识别国家重点实验室_视频内容安全
通讯作者Bing Li
作者单位1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
2.Univ London, Birkbeck Coll, Dept Comp Sci & Informat Syst, London WC1E 7HK, England
3.Natl Univ Singapore, Dept Elect & Comp Engn, Singapore 117583, Singapore
推荐引用方式
GB/T 7714
Li, Bing,Xiong, Weihua,Wu, Ou,et al. Horror Image Recognition Based on Context-Aware Multi-Instance Learning[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2015,24(12):5193-5205.
APA Li, Bing.,Xiong, Weihua.,Wu, Ou.,Hu, Weiming.,Maybank, Stephen.,...&Bing Li.(2015).Horror Image Recognition Based on Context-Aware Multi-Instance Learning.IEEE TRANSACTIONS ON IMAGE PROCESSING,24(12),5193-5205.
MLA Li, Bing,et al."Horror Image Recognition Based on Context-Aware Multi-Instance Learning".IEEE TRANSACTIONS ON IMAGE PROCESSING 24.12(2015):5193-5205.
条目包含的文件 下载所有文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
bingli-tip.pdf(906KB)期刊论文作者接受稿开放获取CC BY-NC-SA浏览 下载
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[Li, Bing]的文章
[Xiong, Weihua]的文章
[Wu, Ou]的文章
百度学术
百度学术中相似的文章
[Li, Bing]的文章
[Xiong, Weihua]的文章
[Wu, Ou]的文章
必应学术
必应学术中相似的文章
[Li, Bing]的文章
[Xiong, Weihua]的文章
[Wu, Ou]的文章
相关权益政策
暂无数据
收藏/分享
文件名: bingli-tip.pdf
格式: Adobe PDF
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。