Component Awareness in Convolutional Neural Networks

Brian Kenji Iwana, Letao Zhou, Kumiko Tanaka-Ishii, Seiichi Uchida

研究成果: Chapter in Book/Report/Conference proceedingConference contribution

抄録

In this work, we investigate the ability of Convolutional Neural Networks (CNN) to infer the presence of components that comprise an image. In recent years, CNNs have achieved powerful results in classification, detection, and segmentation. However, these models learn from instance-level supervision of the detected object. In this paper, we determine if CNNs can detect objects using image-level weakly supervised labels without localization. To demonstrate that a CNN can infer awareness of objects, we evaluate a CNN's classification ability with a database constructed of Chinese characters with only character-level labeled components. We show that the CNN is able to achieve a high accuracy in identifying the presence of these components without specific knowledge of the component. Furthermore, we verify that the CNN is deducing the knowledge of the target component by comparing the results to an experiment with the component removed. This research is important for applications with large amounts of data without robust annotation such as Chinese character recognition.

本文言語英語
ホスト出版物のタイトルProceedings - 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017
出版社IEEE Computer Society
ページ394-399
ページ数6
ISBN(電子版)9781538635865
DOI
出版ステータス出版済み - 1 25 2018
イベント14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017 - Kyoto, 日本
継続期間: 11 9 201711 15 2017

出版物シリーズ

名前Proceedings of the International Conference on Document Analysis and Recognition, ICDAR
1
ISSN(印刷版)1520-5363

その他

その他14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017
Country日本
CityKyoto
Period11/9/1711/15/17

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

フィンガープリント 「Component Awareness in Convolutional Neural Networks」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル