Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (6): 1593-1600.DOI: 10.11772/j.issn.1001-9081.2019101774

• Artificial intelligence • Previous Articles     Next Articles

Automatic annotation of visual deep neural network

LI Ming1,2, GUO Chenhao1,2, CHEN Xing1,2   

  1. 1. College of Mathematics and Computer Science, Fuzhou University, Fuzhou Fujian 350108, China
    2. Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing (Fuzhou University), Fuzhou Fujian 350108, China
  • Received:2019-10-19 Revised:2019-12-30 Online:2020-06-10 Published:2020-06-18
  • Contact: CHEN Xing, born in 1985, Ph. D., associate professor. His research interests include system software, software self-adaptation, cloud computing.
  • About author:LI Ming, born in 1997, M. S. candidate. His research interests include DNN services composition.GUO Chenhao, born in 1997. His research interests include machine learning, natural language processing.CHEN Xing, born in 1985, Ph. D., associate professor. His research interests include system software, software self-adaptation, cloud computing.
  • Supported by:
    National Key Research and Development Program of China (2018YFB1004800), the Talent Program of Fujian Province for Distinguished Young Scholars in Higher Education, the Guiding Project of Fujian Province (2018H0017).

视觉类深度神经网络的自动标注

李鸣1,2, 郭晨皓1,2, 陈星1,2   

  1. 1.福州大学 数学与计算机科学学院,福州 350108
    2.福建省网络计算与智能信息处理重点实验室(福州大学),福州 350108
  • 通讯作者: 陈星(1985—)
  • 作者简介:李鸣(1997—),男,福建莆田人,硕士研究生,主要研究方向:DNN服务组装。郭晨皓(1997—),男,福建福州人,主要研究方向:机器学习、自然语言处理。陈星(1985—),男,福建福州人,副教授,博士,主要研究方向:系统软件、软件自适应、云计算。
  • 基金资助:
    国家重点研发计划项目(2018YFB1004800);福建省高校杰出青年科研人才计划项目;福建省引导性项目(2018H0017)。

Abstract: Focused on the issue that developers cannot quickly figure out the models they need from various models, an automatic annotation method of visual deep neural network based on natural language processing technology was proposed. Firstly, the field categories of visual neural networks were divided, the keywords and corresponding weights were calculated according to the word frequency and other information. Secondly, a keyword extractor was established to extract keywords from paper abstracts. Finally, the similarities between extracted keywords and the known weights were calculated in order to obtain the application fields of a specific model. With experimental data derived from the papers published in three top international conferences of computer vision: IEEE International Conference on Computer Vision(ICCV), IEEE Conference on Computer Vision and Pattern Recognition(CVPR) and European Conference on Computer Vision(ECCV), the experiments were carried out. The experimental results indicate that the proposed method provides highly accurate classification results with a macro average value of 0.89. The validity of this proposed method is verified.

Key words: computer vision, deep neural network, text classification, keyword extraction, automatic annotation, model application field

摘要: 针对开发人员难以快速从众多模型中找到自己所需的模型的问题,提出了一种基于自然语言处理技术的视觉类深度神经网络的自动标注方法。首先,划分视觉类神经网络的领域类别,根据词频等信息计算关键词及其对应的权值;其次,建立关键词提取器从论文摘要中提取出关键词;最后,将提取得到的关键词和已知权值进行相似度计算,从而得到模型的应用领域。从三大国际计算机视觉领域会议,即国际计算机视觉大会(ICCV)、IEEE国际计算机视觉与模式识别会议(CVPR)和欧洲计算机视觉国际会议(ECCV)发表的论文中选取实验数据进行实验。实验结果表明,所提方法能够提供宏平均值为0.89的高精度分类结果,验证了该方法的有效性。

关键词: 计算机视觉, 深度神经网络, 文本分类, 关键词提取, 自动标注, 模型应用领域

CLC Number: