An Effective SVM-based Active Feedback Framework for Image Retrieval

被引：0

作者：

Rao, Yunbo ^{[1
]}

Liu, Wei ^{[1
]}

Wang, Shiqi ^{[1
]}

Song, Jiali ^{[1
]}

Fan, Bojiang ^{[1
]}

Gou, Jianping ^{[2
]}

He, Wu ^{[3
]}

机构：

[1] Univ Elect Sci & Technol China, Sch Informat & Software Engn, Chengdu 610054, Peoples R China

[2] Jiangsu Univ, Sch Comp Sci & Telecommun Engn, Zhenjiang 212013, Peoples R China

[3] Sichuan Normal Univ, Coll Movie & Media, Chengdu 610068, Sichuan, Peoples R China

来源：

2017 INTERNATIONAL CONFERENCE ON SECURITY, PATTERN ANALYSIS, AND CYBERNETICS (SPAC) | 2017年

关键词：

Relevance feedback; Image retrieval; SVM; Clustering analysis; RELEVANCE FEEDBACK;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Due to the time and hardware restriction, the amount of feedback information is limited in each image retrieval loop. To solve this problem, this paper proposes an effective relevance feedback (RF) method based on Support Vector Machine (SVM) framework, which increases the amount of feedback information by cluster analysis and utilizing unlabeled images to build SVM classifier. As a result, a pseudo-label strategy, consist of a feature subspace partition algorithm and a cluster analysis scheme, is proposed for unlabeled images selection. Experimental results demonstrate the relative high effectiveness of our proposed active framework.

引用

页码：228 / 231

页数：4

共 17 条

[1] A Novel Active Learning Method in Relevance Feedback for Content-Based Remote Sensing Image Retrieval [J].

Demir, Beguem ;

Bruzzone, Lorenzo .

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2015, 53 (05) :2323-2334

[2] Iterative Quantization: A Procrustean Approach to Learning Binary Codes for Large-Scale Image Retrieval [J].

Gong, Yunchao ;

Lazebnik, Svetlana ;

Gordo, Albert ;

Perronnin, Florent .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2013, 35 (12) :2916-2929

[3] An active learning-based SVM multi-class classification model [J].

Guo, Husheng ;

Wang, Wenjian .

PATTERN RECOGNITION, 2015, 48 (05) :1577-1597

[4] Connected Component Model for Multi-Object Tracking [J].

He, Zhenyu ;

Li, Xin ;

You, Xinge ;

Tao, Dacheng ;

Tang, Yuan Yan .

IEEE TRANSACTIONS ON IMAGE PROCESSING, 2016, 25 (08) :3698-3711

[5] Semisupervised SVM Batch Mode Active Learning with Applications to Image Retrieval [J].

Hoi, Steven C. H. ;

Jin, Rong ;

Zhu, Jianke ;

Lyu, Michael R. .

ACM TRANSACTIONS ON INFORMATION SYSTEMS, 2009, 27 (03)

[6]

Jegou H, 2008, LECT NOTES COMPUT SC, V5302, P304, DOI 10.1007/978-3-540-88682-2_24

[7] Combining active learning and semi-supervised learning to construct SVM classifier [J].

Leng, Yan ;

Xu, Xinyan ;

Qi, Guanghui .

KNOWLEDGE-BASED SYSTEMS, 2013, 44 :121-131

[8] Automatic linguistic indexing of pictures by a statistical modeling approach [J].

Li, J ;

Wang, JZ .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2003, 25 (09) :1075-1088

[9]

Li J., 2015, HDB NEURAL INFO PROC, P433

[10] SVM-based active feedback in image retrieval using clustering and unlabeled data [J].

Liu, Rujie ;

Wanga, Yuehong ;

Baba, Takayuki ;

Masumoto, Daiki ;

Nagata, Shigemi .

PATTERN RECOGNITION, 2008, 41 (08) :2645-2655

← 1 2 →