Article(id=1242137057234068329, tenantId=1146029695717560320, journalId=1146031591421210625, issueId=1242137051269772083, articleNumber=null, orderNo=5, doi=10.3981/j.issn.1000-7857.2018.17.003, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1531584000000, receivedDateStr=2018-07-15, revisedDate=1534262400000, revisedDateStr=2018-08-15, acceptedDate=null, acceptedDateStr=null, onlineDate=1537233511637, onlineDateStr=2018-09-18, pubDate=1536768000000, pubDateStr=2018-09-13, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1537233511637, onlineIssueDateStr=2018-09-18, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1774078152231, creator=sys-migrate, updateTime=1774078152231, updator=sys-migrate, issue=Issue{id=1242137051269772083, tenantId=1146029695717560320, journalId=1146031591421210625, year='2018', volume='36', issue='17', pageStart='1', pageEnd='112', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=3, issueType=-1, specialIssue=null, createTime=1774078150809, creator=sys-migrate, updateTime=1774078150809, updator=sys-migrate, preIssue=null, nextIssue=null, ext=null, issueFiles=null}, startPage=19, endPage=22, ext={EN=ArticleExt(id=1242137062506308494, articleId=1242137057234068329, tenantId=1146029695717560320, journalId=1146031591421210625, language=EN, title=From simulation to real, adversarial learning based data augmention method, columnId=1242136927416168874, journalTitle=Science & Technology Review, columnName=Scientific Comments, runingTitle=null, highlight=null, articleAbstract=Deep learning has recently made a huge breakthrough in the field of computer vision. What makes it succeed is using a large amount of labeled data for supervised learning with deep neural networks. However, labeling a large-scale dataset is very expensive and time-consuming. To solve the large-scale dataset annotation issue, Apple's Shrivastava team tried to achieve unsupervised learning of simulated images with existing computer simulation techniques and adversarial training methods, thereby avoiding the expensive image annotation process. They had three innovations, namely a ‘self-regularization’ term, a local adversarial loss, and updating the discriminator using a history of refined images so that the real image is generated while retaining the input image features. The experiment results showed that the method can generate highly realistic images. The team also quantitatively analyzed the generated images by training a gaze estimation model and a hand posture estimation model. The results indicated a significant improvement over using synthetic images and achieved the state of the art on the MPⅡGaze dataset without any labeled real data. However, the researchers didn't conduct any experiment in complex scenarios involving multiple objects. The application of the proposed method still has limitations in complex scenarios., correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=Gfq/i8z6hzPkjKxej8GSEQ==, pdfFileSize=1312121, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou 310027, China, fund=null, authors=LIU Yong, ZENG Xianfang, authorsList=LIU Yong, ZENG Xianfang), CN=ArticleExt(id=1242137061533233994, articleId=1242137057234068329, tenantId=1146029695717560320, journalId=1146031591421210625, language=CN, title=“弄假成真”:基于对抗学习的数据增广方法, columnId=1150494642043760854, journalTitle=科技导报, columnName=科技评论, runingTitle=null, highlight=null, articleAbstract=近年来,深度学习在计算机视觉领域取得了巨大的突破,其背后是利用大量标签数据对深度网络进行监督训练,而标注大规模数据集非常昂贵且十分耗时。针对大规模数据集标注问题,苹果公司的Shrivastava团队希望借助现有的计算机仿真技术以及对抗训练的方法,实现仿真图像的无监督学习,从而避免昂贵的图像标注过程。该团队在对抗网络的基础上提出3个创新点:(1)自正则项;(2)局部对抗损失;(3)使用历史生成图片更新判别器,使得生成真实化图片的同时保留输入图像特征。实验结果展示该方法可以生成高度真实化的图片。研究者通过训练凝视估计模型、手部姿态估计模型定量分析生成图片的效果,分析结果表明,使用生成图片训练的模型,在MPⅡGaze数据集上测试效果有很大的提升,达到了当时最好的效果。不过,研究者并未在包含多个物体的复杂场景下进行实验,文中提出的方法在复杂场景下的应用还存在局限性。, correspAuthors=null, authorNote=刘勇,教授,研究方向为机器学习、机器人视觉,电子信箱:Yongliu@iipc.zju.edu.cn;曾仙芳(共同第一作者),博士研究生,研究方向为计算机视觉,电子信箱:zzlongjuanfeng@zju.edu.cn, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=ZcgAsh74oCqtE5+I7us3mA==, pdfFileSize=1312121, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=浙江大学智能系统与控制研究所, 杭州 310027, fund=null, authors=刘勇, 曾仙芳, authorsList=刘勇, 曾仙芳)}, authors=[Author(id=1276790283492782822, tenantId=1146029695717560320, journalId=null, articleId=1242137057234068329, orderNo=null, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=null, email=null, emailSecond=null, emailThird=null, correspondingAuthor=null, authorType=null, ext={EN=AuthorExt(id=null, tenantId=null, journalId=1146031591421210625, articleId=1242137057234068329, authorId=1276790283492782822, language=EN, stringName=LIU Yong, ZENG Xianfang, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=null, tenantId=null, journalId=1146031591421210625, articleId=1242137057234068329, authorId=1276790283492782822, language=CN, stringName=刘勇, 曾仙芳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=null)], keywords=[Keyword(id=1242137059809375043, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1242137057234068329, language=CN, orderNo=1, keyword=对抗训练), Keyword(id=1242137059914232645, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1242137057234068329, language=CN, orderNo=1, keyword=无监督学习), Keyword(id=1242137060006507336, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1242137057234068329, language=CN, orderNo=1, keyword=仿真图片), Keyword(id=1242137060094587721, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1242137057234068329, language=CN, orderNo=1, keyword=深度学习), Keyword(id=1242137062099461002, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1242137057234068329, language=EN, orderNo=1, keyword=adversarial training), Keyword(id=1242137062179152779, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1242137057234068329, language=EN, orderNo=1, keyword=unsupervised learning), Keyword(id=1242137062258844556, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1242137057234068329, language=EN, orderNo=1, keyword=simulated image), Keyword(id=1242137062338536333, tenantId=1146029695717560320, journalId=1146031591421210625, articleId=1242137057234068329, language=EN, orderNo=1, keyword=deep learning)], refs=null, funds=null, companyList=null, figs=null, attaches=null, journal=Journal(id=1125356956822126595, delFlag=0, nameCn=科技导报, nameEn=Science & Technology Review, nameHistory1=null, nameHistory2=null, issn=1000-7857, eissn=, cn=11-1421/N, coden=null, periodic=3, language=CN, oaType=0, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=aEuqdCNQUjPEKa3rm5A/8Q==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Rev, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754267492363, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=aEuqdCNQUjPEKa3rm5A/8Q==, picEn=4AIQ9/oc3H8lvjeELJ6WWw==, jcr=null, cjcr=null, exts=[JournalExt(id=1159045127382855686, language=CN, name=科技导报, nameHistory1=null, nameHistory2=null, managedBy=中国科学技术协会, sponsoredBy=中国科学技术协会, publishedBy=科技导报社, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.kjdb.org/CN/home, createdTime=1754267492385, updatedTime=1754267492385, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.kjdb.org/CN/column/column7.shtml, submissionAuthorUrl=https://kjdbauthor.cast.org.cn/webm, submissionEditorUrl=https://kjdbeditor.cast.org.cn/webm/, submissionReviewUrl=https://kjdbauthor.cast.org.cn/webm, submissionCeEditorUrl=https://kjdbeditor.cast.org.cn/webm/, submissionAeEditorUrl=https://kjdbeditor.cast.org.cn/webm/, option={"copyright":""}), JournalExt(id=1159045127433187335, language=EN, name=Science & Technology Review, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.kjdb.org/EN/home, createdTime=1754267492398, updatedTime=1754267492398, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.kjdb.org/EN/column/column7.shtml, submissionAuthorUrl=https://kjdbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://kjdbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://kjdbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://kjdbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://kjdbeditor.manuscriptcloud.com/login, option={"copyright":""})], databaseList=null, tenantJournalId=1146031591421210625, websiteList=[Website(id=1146104741081231361, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031591421210625, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kjdb/CN, language=CN, createTime=1751182263881, createBy=18614031015, updateTime=1751778001962, updateBy=18614031015, name=科技导报, tplId=1146099689490845704, title=科技导报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148021146403992296, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=articleTextType, value=kx, createTime=1751639170504, updateTime=1751639170504, creator=18614031015, updator=18614031015), WebsiteProps(id=1148021146378826469, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=banner, value=null, createTime=1751639170498, updateTime=1751639170498, creator=18614031015, updator=18614031015), WebsiteProps(id=1148021146366243556, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=9GHSf7eGlIPH0Tv/OOdstA==, createTime=1751639170495, updateTime=1751639170495, creator=18614031015, updator=18614031015), WebsiteProps(id=1148021146395603687, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751639170502, updateTime=1751639170502, creator=18614031015, updator=18614031015), WebsiteProps(id=1148021146387215078, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146104741081231361, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751639170500, updateTime=1751639170500, creator=18614031015, updator=18614031015)]), Website(id=1146105254833139715, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031591421210625, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kjdb/EN, language=EN, createTime=1751182386363, createBy=18614031015, updateTime=1753500121937, updateBy=18614031015, name=科技导报, tplId=1146101810881728533, title=Science & Technology Review, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155838567709528217, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=articleTextType, value=kx, createTime=1753502988984, updateTime=1753502988984, creator=18614031015, updator=18614031015), WebsiteProps(id=1155838567692750998, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=banner, value=null, createTime=1753502988980, updateTime=1753502988980, creator=18614031015, updator=18614031015), WebsiteProps(id=1155838567688556693, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/EN/file/pic?fileId=9GHSf7eGlIPH0Tv/OOdstA==, createTime=1753502988979, updateTime=1753502988979, creator=18614031015, updator=18614031015), WebsiteProps(id=1155838567705333912, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/EN/file/pic, createTime=1753502988983, updateTime=1753502988983, creator=18614031015, updator=18614031015), WebsiteProps(id=1155838567701139607, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146105254833139715, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1753502988982, updateTime=1753502988982, creator=18614031015, updator=18614031015)])], journalTitle=科技导报, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science & Technology Review, journalPhotoCn=aEuqdCNQUjPEKa3rm5A/8Q==, journalPhotoEn=4AIQ9/oc3H8lvjeELJ6WWw==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=1, jcrJf=null, cjcrJf=0.91, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kjdb/CN/10.3981/j.issn.1000-7857.2018.17.003, detailUrlEn=https://castjournals.cast.org.cn/joweb/kjdb/EN/10.3981/j.issn.1000-7857.2018.17.003, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kjdb/CN/PDF/10.3981/j.issn.1000-7857.2018.17.003, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kjdb/EN/PDF/10.3981/j.issn.1000-7857.2018.17.003, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
“弄假成真”:基于对抗学习的数据增广方法
收藏切换
PDF下载
刘勇, 曾仙芳
科技导报 | 科技评论 2018,36(17): 19-22
收起
收藏切换
科技导报 | 科技评论 2018, 36(17): 19-22
“弄假成真”:基于对抗学习的数据增广方法
全屏
刘勇, 曾仙芳
作者信息
From simulation to real, adversarial learning based data augmention method
LIU Yong, ZENG Xianfang
Affiliations
出版时间: 2018-09-13 doi: 10.3981/j.issn.1000-7857.2018.17.003
文章导航
收藏切换
近年来,深度学习在计算机视觉领域取得了巨大的突破,其背后是利用大量标签数据对深度网络进行监督训练,而标注大规模数据集非常昂贵且十分耗时。针对大规模数据集标注问题,苹果公司的Shrivastava团队希望借助现有的计算机仿真技术以及对抗训练的方法,实现仿真图像的无监督学习,从而避免昂贵的图像标注过程。该团队在对抗网络的基础上提出3个创新点:(1)自正则项;(2)局部对抗损失;(3)使用历史生成图片更新判别器,使得生成真实化图片的同时保留输入图像特征。实验结果展示该方法可以生成高度真实化的图片。研究者通过训练凝视估计模型、手部姿态估计模型定量分析生成图片的效果,分析结果表明,使用生成图片训练的模型,在MPⅡGaze数据集上测试效果有很大的提升,达到了当时最好的效果。不过,研究者并未在包含多个物体的复杂场景下进行实验,文中提出的方法在复杂场景下的应用还存在局限性。
对抗训练  /  无监督学习  /  仿真图片  /  深度学习
Deep learning has recently made a huge breakthrough in the field of computer vision. What makes it succeed is using a large amount of labeled data for supervised learning with deep neural networks. However, labeling a large-scale dataset is very expensive and time-consuming. To solve the large-scale dataset annotation issue, Apple's Shrivastava team tried to achieve unsupervised learning of simulated images with existing computer simulation techniques and adversarial training methods, thereby avoiding the expensive image annotation process. They had three innovations, namely a ‘self-regularization’ term, a local adversarial loss, and updating the discriminator using a history of refined images so that the real image is generated while retaining the input image features. The experiment results showed that the method can generate highly realistic images. The team also quantitatively analyzed the generated images by training a gaze estimation model and a hand posture estimation model. The results indicated a significant improvement over using synthetic images and achieved the state of the art on the MPⅡGaze dataset without any labeled real data. However, the researchers didn't conduct any experiment in complex scenarios involving multiple objects. The application of the proposed method still has limitations in complex scenarios.
adversarial training  /  unsupervised learning  /  simulated image  /  deep learning
刘勇, 曾仙芳. “弄假成真”:基于对抗学习的数据增广方法. 科技导报, 2018 , 36 (17) : 19 -22 . DOI: 10.3981/j.issn.1000-7857.2018.17.003
LIU Yong, ZENG Xianfang. From simulation to real, adversarial learning based data augmention method[J]. Science & Technology Review, 2018 , 36 (17) : 19 -22 . DOI: 10.3981/j.issn.1000-7857.2018.17.003
2018年第36卷第17期
PDF下载
247
95
引用本文
BibTeX
文章信息
doi: 10.3981/j.issn.1000-7857.2018.17.003
  • 接收时间:2018-07-15
  • 首发时间:2018-09-18
  • 出版时间:2018-09-13
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2018-07-15
  • 修回日期:2018-08-15
基金
作者信息
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kjdb/CN/10.3981/j.issn.1000-7857.2018.17.003
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏