Multi-label active learning for image classification

Jian Wu; Victor S. Sheng; Jing Zhang; Pengpeng Zhao; Zhiming Cui

doi:10.1109/ICIP.2014.7026058

Multi-label active learning for image classification

Jian Wu, Victor S. Sheng, Jing Zhang, Pengpeng Zhao, Zhiming Cui

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

23 Scopus citations

Abstract

Multi-label image data is becoming ubiquitous. Image semantic understanding is typically formulated as a classification problem. This paper focuses on multi-label active learning for image classification. It first extends a traditional example based active learning method for multilabel active learning for image classification. Since the traditional example based active method doesn't work well, we propose a novel example-label based multi-label active learning method. Our experimental results on two image datasets demonstrate that the proposed method significantly reduces the labeling workload and improves the performance of the built classifier. Additionally, we conduct experiments on two other types of multi-label datasets for validating the versatility of our proposed method, and the experimental results show the consistent effect.

Original language	English
Title of host publication	2014 IEEE International Conference on Image Processing, ICIP 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	5227-5231
Number of pages	5
ISBN (Electronic)	9781479957514
DOIs	https://doi.org/10.1109/ICIP.2014.7026058
State	Published - Jan 28 2014

Publication series

Name	2014 IEEE International Conference on Image Processing, ICIP 2014

Keywords

Multi-label
active learning
example-label pair
image classification

Access to Document

10.1109/ICIP.2014.7026058

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Multi-label active learning for image classification'. Together they form a unique fingerprint.

Cite this

Wu, J., Sheng, V. S., Zhang, J., Zhao, P., & Cui, Z. (2014). Multi-label active learning for image classification. In 2014 IEEE International Conference on Image Processing, ICIP 2014 (pp. 5227-5231). [7026058] (2014 IEEE International Conference on Image Processing, ICIP 2014). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICIP.2014.7026058

@inproceedings{4098e235e3844e0696fbcb4e5dbbca3d,

title = "Multi-label active learning for image classification",

abstract = "Multi-label image data is becoming ubiquitous. Image semantic understanding is typically formulated as a classification problem. This paper focuses on multi-label active learning for image classification. It first extends a traditional example based active learning method for multilabel active learning for image classification. Since the traditional example based active method doesn't work well, we propose a novel example-label based multi-label active learning method. Our experimental results on two image datasets demonstrate that the proposed method significantly reduces the labeling workload and improves the performance of the built classifier. Additionally, we conduct experiments on two other types of multi-label datasets for validating the versatility of our proposed method, and the experimental results show the consistent effect.",

keywords = "Multi-label, active learning, example-label pair, image classification",

author = "Jian Wu and Sheng, {Victor S.} and Jing Zhang and Pengpeng Zhao and Zhiming Cui",

year = "2014",

month = jan,

day = "28",

doi = "10.1109/ICIP.2014.7026058",

language = "English",

series = "2014 IEEE International Conference on Image Processing, ICIP 2014",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "5227--5231",

booktitle = "2014 IEEE International Conference on Image Processing, ICIP 2014",

}

Multi-label active learning for image classification. / Wu, Jian; Sheng, Victor S.; Zhang, Jing; Zhao, Pengpeng; Cui, Zhiming.

2014 IEEE International Conference on Image Processing, ICIP 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 5227-5231 7026058 (2014 IEEE International Conference on Image Processing, ICIP 2014).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Multi-label active learning for image classification

AU - Wu, Jian

AU - Sheng, Victor S.

AU - Zhang, Jing

AU - Zhao, Pengpeng

AU - Cui, Zhiming

PY - 2014/1/28

Y1 - 2014/1/28

N2 - Multi-label image data is becoming ubiquitous. Image semantic understanding is typically formulated as a classification problem. This paper focuses on multi-label active learning for image classification. It first extends a traditional example based active learning method for multilabel active learning for image classification. Since the traditional example based active method doesn't work well, we propose a novel example-label based multi-label active learning method. Our experimental results on two image datasets demonstrate that the proposed method significantly reduces the labeling workload and improves the performance of the built classifier. Additionally, we conduct experiments on two other types of multi-label datasets for validating the versatility of our proposed method, and the experimental results show the consistent effect.

AB - Multi-label image data is becoming ubiquitous. Image semantic understanding is typically formulated as a classification problem. This paper focuses on multi-label active learning for image classification. It first extends a traditional example based active learning method for multilabel active learning for image classification. Since the traditional example based active method doesn't work well, we propose a novel example-label based multi-label active learning method. Our experimental results on two image datasets demonstrate that the proposed method significantly reduces the labeling workload and improves the performance of the built classifier. Additionally, we conduct experiments on two other types of multi-label datasets for validating the versatility of our proposed method, and the experimental results show the consistent effect.

KW - Multi-label

KW - active learning

KW - example-label pair

KW - image classification

UR - http://www.scopus.com/inward/record.url?scp=84949927263&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2014.7026058

DO - 10.1109/ICIP.2014.7026058

M3 - Conference contribution

AN - SCOPUS:84949927263

T3 - 2014 IEEE International Conference on Image Processing, ICIP 2014

SP - 5227

EP - 5231

BT - 2014 IEEE International Conference on Image Processing, ICIP 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -