Food Category Transfer with Conditional CycleGAN and a Large-Scale Food Image Dataset

Daichi Horita, Ryosuke Tanno, Wataru Shimoda, Keiji Yanai

IJCAI 2018 pp. 67-70

doi:10.1145/3230519.3230597 /ijcai/2018/horita2018ijcai-food/

Abstract

This paper describes "food image transformation" based on a conditional cycleGAN[A3] (cCycleGAN) with a large-scale food image data collected from the Twitter stream. A cCycleGAN is an extension of CycleGAN, which enables "food category transfer" among 10 types of foods and retain the shape of a given food. We experimentally show that 200 and 30,000 food images with the cCycleGAN enable a very natural food category transfer among 10 types of typical Japanese foods: ramen noodle, curry rice, fried rice, beef rice bowl, chilled noodle, spaghetti with meat source, white rice, eel bowl, and fried noodle.

IJCAI Semantic Scholar

Cite

Text

Horita et al. "Food Category Transfer with Conditional CycleGAN and a Large-Scale Food Image Dataset." International Joint Conference on Artificial Intelligence, 2018. doi:10.1145/3230519.3230597

Markdown

[Horita et al. "Food Category Transfer with Conditional CycleGAN and a Large-Scale Food Image Dataset." International Joint Conference on Artificial Intelligence, 2018.](https://mlanthology.org/ijcai/2018/horita2018ijcai-food/) doi:10.1145/3230519.3230597

BibTeX

@inproceedings{horita2018ijcai-food,
  title     = {{Food Category Transfer with Conditional CycleGAN and a Large-Scale Food Image Dataset}},
  author    = {Horita, Daichi and Tanno, Ryosuke and Shimoda, Wataru and Yanai, Keiji},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2018},
  pages     = {67-70},
  doi       = {10.1145/3230519.3230597},
  url       = {https://mlanthology.org/ijcai/2018/horita2018ijcai-food/}
}