Food category transfer with conditional cycleGAN and a large-scale food image dataset

Abstract

This paper describes 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 category 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.