A Novel Biologically Inspired Structural Model for Feature Correspondence

doi:10.1109/TCDS.2022.3188333

CORC > 自动化研究所 > 中国科学院自动化研究所 > 多模态人工智能系统全国重点实验室

	A Novel Biologically Inspired Structural Model for Feature Correspondence
	Lu, Yan-Feng 2,3; Yang, Xu 2,3; Li, Yi 1; Yu, Qian 3; Liu, Zhi-Yong 2,3; Qiao, Hong 2,3
刊名	IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS
	2023-06-01
卷号	15 期号:2 页码:844-854
关键词	Visualization Biological system modeling Biology Brain modeling Biological information theory Task analysis Strain Appearance feature descriptor biologically inspired model feature correspondence feature representation graph matching (GM) graph structure
ISSN号	2379-8920
DOI	10.1109/TCDS.2022.3188333
通讯作者	Yang, Xu(xu.yang@ia.ac.cn)
英文摘要	Feature correspondence is an essential issue in computer science, which could be well formulated by graph matching (GM). However, traditional GM is susceptible to outliers, thus limiting the applications. To address the issue, we present a biologically inspired feature descriptor (BIFD) corresponding to the simple and complex cell layers in primary visual cortex, which shows robust performance against deformations. Furthermore, we propose a novel biologically inspired structural model (BISM) for feature correspondence by fusing the graph structural information and appearance information described by BIFD in the images. The proposed BIFD imitates the cortical pooling operation across multiscale and multiangle cell layers, which makes BISM robust to outliers and distortions. To demonstrate the validity of the proposed method, we evaluate it in feature correspondence tasks on the public databases. The experimental results on synthetic data prove the validity of the proposed method.
资助项目	National Key Research and Development Plan of China[2020AAA0105900] ; Beijing Natural Science Foundation[L211023] ; National Natural Science Foundation of China[91948303] ; National Natural Science Foundation of China[61973301] ; Youth Innovation Promotion Association CAS
WOS关键词	OBJECT RECOGNITION
WOS研究方向	Computer Science ; Robotics ; Neurosciences & Neurology
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:001005746000046
资助机构	National Key Research and Development Plan of China ; Beijing Natural Science Foundation ; National Natural Science Foundation of China ; Youth Innovation Promotion Association CAS
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/53651]
专题	多模态人工智能系统全国重点实验室
通讯作者	Yang, Xu
作者单位	1.Nanchang Univ, Sch Informat Engn, Nanchang 330031, Peoples R China 2.Chinese Acad Sci, Inst Automat, Natl Key Lab Multimodal Artificial Intelligence Sy, 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	Lu, Yan-Feng,Yang, Xu,Li, Yi,et al. A Novel Biologically Inspired Structural Model for Feature Correspondence[J]. IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS,2023,15(2):844-854.
APA	Lu, Yan-Feng,Yang, Xu,Li, Yi,Yu, Qian,Liu, Zhi-Yong,&Qiao, Hong.(2023).A Novel Biologically Inspired Structural Model for Feature Correspondence.IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS,15(2),844-854.
MLA	Lu, Yan-Feng,et al."A Novel Biologically Inspired Structural Model for Feature Correspondence".IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS 15.2(2023):844-854.