CASIA OpenIR  > 紫东太初大模型研究中心  > 图像与视频分析
A Novel Data Augmentation Scheme for Pedestrian Detection with Attribute Preserving GAN
Liu, Songyan1,2; Guo, Haiyun1,2; Hu, Jian-Guo3; Zhao, Xu1,2; Zhao, Chaoyang1,2; Wang, Tong1,2; Zhu, Yousong1,2; Wang, Jinqiao1,2; Tang, Ming1,2
Source PublicationNeurocomputing
ISSN0925-2312
2020-08-11
Volume401Issue:11Pages:123-132
Abstract

Recently pedestrian detection has progressed significantly. However, detecting pedestrians of small scale or in heavy occlusions is still notoriously difficult. Besides, the generalization ability of pre-trained detectors across different datasets remains to be improved. Both of these issues can be attributed to insufficient training data coverage. To cope with this, we present an efficient data augmentation scheme by transferring pedestrians from other datasets into the target scene with a novel Attribute Preserving Generative Adversarial Networks (APGAN). The proposed methodology consists of two steps: pedestrian embedding and style transfer. The former step can simulate pedestrian images of various scale and occlusion, in any pose or background, thus greatly promoting the data variation. The latter step aims to make the generated samples more realistic while guarantee the data coverage. To achieve this goal, we propose APGAN, which pursues both good visual quality and attribute preserving after style transfer. With the proposed method, we can make effective sample augmentations to improve the generalization ability of the trained detectors and enhance its robustness to scale change and occlusions. Extensive experiment results validate the effectiveness and advantages of our method.

KeywordGenerative Adversarial Networks Pedestrian detection Data augmentation
DOI10.1016/j.neucom.2020.02.094
Indexed BySCI
Language英语
Funding ProjectNational Natural Science Foundation of China[61806200] ; National Natural Science Foundation of China[61772527] ; National Natural Science Foundation of China[61772527] ; National Natural Science Foundation of China[61806200]
WOS Research AreaComputer Science
WOS SubjectComputer Science, Artificial Intelligence
WOS IDWOS:000544725700012
PublisherELSEVIER
Sub direction classification目标检测、跟踪与识别
Citation statistics
Cited Times:25[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/39134
Collection紫东太初大模型研究中心_图像与视频分析
Corresponding AuthorGuo, Haiyun; Hu, Jian-Guo
Affiliation1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, 95 Zhongguancun East Rd, Beijing 100190, Peoples R China
2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
3.Sun Yat-Sen University, Guangzhou 510000, China
First Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Corresponding Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Liu, Songyan,Guo, Haiyun,Hu, Jian-Guo,et al. A Novel Data Augmentation Scheme for Pedestrian Detection with Attribute Preserving GAN[J]. Neurocomputing,2020,401(11):123-132.
APA Liu, Songyan.,Guo, Haiyun.,Hu, Jian-Guo.,Zhao, Xu.,Zhao, Chaoyang.,...&Tang, Ming.(2020).A Novel Data Augmentation Scheme for Pedestrian Detection with Attribute Preserving GAN.Neurocomputing,401(11),123-132.
MLA Liu, Songyan,et al."A Novel Data Augmentation Scheme for Pedestrian Detection with Attribute Preserving GAN".Neurocomputing 401.11(2020):123-132.
Files in This Item: Download All
File Name/Size DocType Version Access License
1-s2.0-S092523122030(2691KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Liu, Songyan]'s Articles
[Guo, Haiyun]'s Articles
[Hu, Jian-Guo]'s Articles
Baidu academic
Similar articles in Baidu academic
[Liu, Songyan]'s Articles
[Guo, Haiyun]'s Articles
[Hu, Jian-Guo]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Liu, Songyan]'s Articles
[Guo, Haiyun]'s Articles
[Hu, Jian-Guo]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 1-s2.0-S0925231220302903-main.pdf
Format: Adobe PDF
This file does not support browsing at this time
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.