“Confidence-aware Practical Anime-style Colorization” by Ishii, Kubo, Shinagawa, Maejima, Funatomi, et al. … – ACM SIGGRAPH HISTORY ARCHIVES

“Confidence-aware Practical Anime-style Colorization” by Ishii, Kubo, Shinagawa, Maejima, Funatomi, et al. …

  • ©

  • ©

  • ©

Conference:


Type(s):


Interest Area:


    Research / Education, AI / Machine Learning, and Rendering

Entry Number: 40

Title:

    Confidence-aware Practical Anime-style Colorization

Session/Category Title:   Image Algorithms


Presenter(s)/Author(s):



Abstract:


    We propose practical anime-style colorization of an input line- drawing. The key idea is the strategic withdrawal which reflects the prediction confidence that indicates the expected accuracy of the predicted color labels. Furthermore, we investigate the relation between the proposed confidence, prediction accuracy, and number of automatically colorized regions to maximize the efficiency of the colorization process including both automatic prediction and manual correction for practical use in production.

References:


    Sophie Ramassamy, Hiroyuki Kubo, Takuya Funatomi, Daichi Ishii, Akinobu Maejima, Satoshi Nakamura, and Yasuhiro Mukaigawa. 2018. Pre- and Post-Processes for Automatic Colorization Using a Fully Convolutional Network. In SIGGRAPH Asia 2018 Posters (SA ’18). Article Article 70, 2 pages.

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. 2015. U-Net: Convolutional Networks for Biomedical Image Segmentation. LNCS 9351, 234–241. https://doi. org/10.1007/978-3-319-24574-4_28

Keyword(s):



Additional Images:

© © © © © © ©

PDF:



ACM Digital Library Publication:



Overview Page:



Submit a story:

If you would like to submit a story about this presentation, please contact us: historyarchives@siggraph.org