Logo Retrieval Using Logo Proposals and Adaptive Weighted Pooling
Qi,Chengzuo; Shi,Cunzhao; Wang,Chunheng; Xiao,Baihua
2017-03-07
发表期刊IEEE Signal Processing Letters
期号4页码:442-445
摘要This letter presents a novel approach for logo re-
trieval.Consideringthefactthatlogoonlyoccupiesasmallportion
of an image, we apply Faster R-CNN to detect logo proposals first,
and then use a two-step pooling strategy with adaptive weight to
obtainanaccurateglobalsignature.Theadaptiveweightedpooling
method can effectively balance the recall and precision of propos-
als by incorporating the probability of each proposal being a logo.
Experimentalresultsshowthattheproposedmethodinterpretsthe
similarity between query and database image more accurately and
achieves state of the art performance.
关键词Adaptive Weighted Pooling Faster R-cnn Logo Retrieval
DOI10.1109/LSP.2017.2673119
引用统计
文献类型期刊论文
条目标识符http://ir.ia.ac.cn/handle/173211/19553
专题复杂系统管理与控制国家重点实验室_影像分析与机器视觉
作者单位Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
推荐引用方式
GB/T 7714
Qi,Chengzuo,Shi,Cunzhao,Wang,Chunheng,et al. Logo Retrieval Using Logo Proposals and Adaptive Weighted Pooling[J]. IEEE Signal Processing Letters,2017(4):442-445.
APA Qi,Chengzuo,Shi,Cunzhao,Wang,Chunheng,&Xiao,Baihua.(2017).Logo Retrieval Using Logo Proposals and Adaptive Weighted Pooling.IEEE Signal Processing Letters(4),442-445.
MLA Qi,Chengzuo,et al."Logo Retrieval Using Logo Proposals and Adaptive Weighted Pooling".IEEE Signal Processing Letters .4(2017):442-445.
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
Logo Retrieval Using(384KB)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
[Qi,Chengzuo]的文章
[Shi,Cunzhao]的文章
[Wang,Chunheng]的文章
百度学术
百度学术中相似的文章
[Qi,Chengzuo]的文章
[Shi,Cunzhao]的文章
[Wang,Chunheng]的文章
必应学术
必应学术中相似的文章
[Qi,Chengzuo]的文章
[Shi,Cunzhao]的文章
[Wang,Chunheng]的文章
相关权益政策
暂无数据
收藏/分享
文件名: Logo Retrieval Using Logo Proposals and Adaptive.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。