中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 复杂系统管理与控制国家重点实验室
  5. 影像分析与机器视觉团队
Selective feature connection mechanism: Concatenating multi-layer CNN features with a feature selector

文献类型：期刊论文

作者Du, Chen1,2; Wang, Chunheng2; Wang, Yanna2; Shi, Cunzhao2; Xiao, Baihua2
刊名PATTERN RECOGNITION LETTERS
出版日期2020
卷号129页码:108-114
关键词Feature combination Network architecture Selective feature connection mechanism Convolutional neural network
ISSN号0167-8655
DOI10.1016/j.patrec.2019.11.015
通讯作者Wang, Yanna(chunheng.wang@ia.ac.cn)
英文摘要Different layers of deep convolutional neural networks(CNNs) can encode different-level information. High-layer features always contain more semantic information, and low-layer features contain more detail information. However, low-layer features suffer from the background clutter and semantic ambiguity. During visual recognition, the feature combination of the low-layer and high-level features plays an important role in context modulation. If directly combining the high-layer and low-layer features, the background clutter and semantic ambiguity may be caused due to the introduction of detailed information. In this paper, we propose a general network architecture to concatenate CNN features of different layers in a simple and effective way, called Selective Feature Connection Mechanism (SFCM). Low-level features are selectively linked to high-level features with a feature selector which is generated by high-level features. The proposed connection mechanism can effectively overcome the above-mentioned drawbacks. We demonstrate the effectiveness, superiority, and universal applicability of this method on multiple challenging computer vision tasks, including image classification, scene text detection, and image-to-image translation. (C) 2019 Elsevier B.V. All rights reserved.
资助项目Key Programs of the Chinese Academy of Sciences[ZDBS-SSWJSC003] ; Key Programs of the Chinese Academy of Sciences[ZDBS-SSW-JSC004] ; Key Programs of the Chinese Academy of Sciences[ZDBS-SSWJSC005] ; National Natural Science Foundation of China (NSFC)[61601462] ; National Natural Science Foundation of China (NSFC)[61531019] ; National Natural Science Foundation of China (NSFC)[71621002]
WOS研究方向Computer Science
语种英语
WOS记录号WOS:000504641500016
出版者ELSEVIER
资助机构Key Programs of the Chinese Academy of Sciences ; National Natural Science Foundation of China (NSFC)
源URL[http://ir.ia.ac.cn/handle/173211/29461]  
专题自动化研究所_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团队
通讯作者Wang, Yanna
作者单位1.Univ Chinese Acad Sci, Beijing, Peoples R China
2.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		 Du, Chen,Wang, Chunheng,Wang, Yanna,et al. Selective feature connection mechanism: Concatenating multi-layer CNN features with a feature selector[J]. PATTERN RECOGNITION LETTERS,2020,129:108-114.
APA Du, Chen,Wang, Chunheng,Wang, Yanna,Shi, Cunzhao,&Xiao, Baihua.(2020).Selective feature connection mechanism: Concatenating multi-layer CNN features with a feature selector.PATTERN RECOGNITION LETTERS,129,108-114.
MLA Du, Chen,et al."Selective feature connection mechanism: Concatenating multi-layer CNN features with a feature selector".PATTERN RECOGNITION LETTERS 129(2020):108-114.

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。