Classification of photo and sketch images using convolutional neural networks

Kazuma Sasaki*, Madoka Yamakawa, Kana Sekiguchi, Tetsuya Ogata

*この研究の対応する著者

研究成果: Conference contribution

3 被引用数 (Scopus)

抄録

Content-Based Image Retrieval (CBIR) system enables us to access images using only images as queries, instead of keywords. Photorealistic images, and hand-drawn sketch image can be used as a queries as well. Recently, convolutional neural networks (CNNs) are used to discriminate images including sketches. However, the tasks are limited to classifying only one type of images, either photo or sketch images, due to the lack of a large dataset of sketch images and the large difference of their visual characteristics. In this paper, we introduce a simple way to prepare training datasets, which can enable the CNN model to classify both types of images by color transforming photo and illustration images. Through the training experiment, we show that the proposed method contributes to the improvement of classification accuracy.

本文言語English
ホスト出版物のタイトルArtificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings
編集者Alessandro E.P. Villa, Paolo Masulli, Antonio Javier Pons Rivero
出版社Springer Verlag
ページ283-290
ページ数8
ISBN(印刷版)9783319447803
DOI
出版ステータスPublished - 2016
イベント25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016 - Barcelona, Spain
継続期間: 2016 9 62016 9 9

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
9887 LNCS
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Other

Other25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016
国/地域Spain
CityBarcelona
Period16/9/616/9/9

ASJC Scopus subject areas

  • 理論的コンピュータサイエンス
  • コンピュータ サイエンス(全般)

フィンガープリント

「Classification of photo and sketch images using 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