期刊论文、会议论文

Ding, Wan

英文

2017 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC 2017)

2309-9402

2017

1368-1372

9th Annual Summit and Conference of the Asia-Pacific-Signal-and-Information-Processing-Association (APSIPA ASC)

Asia-Pacific Signal and Information Processing Association Annual Summit and Conference

DEC 12-15, 2017

Kuala Lumpur, MALAYSIA

[Ding, Wan;Yu, Xinguo] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan, Hubei, Peoples R China.^[Huang, Dong-Yan] Inst Infocomm Res, Human Language Technol Dept, Singapore, Singapore.^[Chen, Zhuo;Lin, Weisi] Nanyang Technol Univ, Sch Comp Sci & Engn, Singapore, Singapore.

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

978-1-5386-1542-3

10.1109/APSIPA.2017.8282246

本校为第一且通讯机构

Positive samples of facial actions are much fewer than negative samples in natural conditions. The highly imbalanced class-distributions may cause very slow rate of convergence of error when using neural networks for facial action recognition. Traditional methods tackle this class-imbalance problem by changing data distributions, which is challenging for preventing the loss of useful information. In this paper we tackle this problem by using very deep (>10 layers) architectures to increase the chance that network training has acceptable rate of convergence using highly imbalanced data sets. Ex...