Help us understand the problem. What is going on with this article?

CycleGANを用いたスタイル変換

More than 1 year has passed since last update.

2018年9月15日の 機械学習名古屋 第17回勉強会 で話した内容をまとめておきます。
内容は CycleGAN 使って遊んだという話です。そんなに上手くいかなかったけど供養として。

CycleGAN の keras 実装は https://github.com/kiyohiro8/cycleGAN_keras
CycleGAN の問題設定や損失設計については CycleGAN にまとめました。

Unpaired Image-to-image translation

image.png

CycleGAN の構造

image.png

Adversarial Loss

image.png

Adversarial Loss だけでは不十分

image.png

image.png

Cycle Consistency Loss

image.png

Fully Objective

image.png

例1 チーズケーキ ⇔ チョコレートケーキ

image.png

image.png

image.png

image.png

例2 風景写真 ⇔ Magic: the Gathering のイラスト

image.png

image.png

image.png

フルサイズの画像に対する変換例 (風景写真 → イラスト)

全体として暗く妖しい雰囲気になってしまいます。
000000.jpg
000000.jpg.png
000022.jpg
000022.jpg.png

Why not register and get more from Qiita?
  1. We will deliver articles that match you
    By following users and tags, you can catch up information on technical fields that you are interested in as a whole
  2. you can read useful information later efficiently
    By "stocking" the articles you like, you can search right away
Comments
No comments
Sign up for free and join this conversation.
If you already have a Qiita account
Why do not you register as a user and use Qiita more conveniently?
You need to log in to use this function. Qiita can be used more conveniently after logging in.
You seem to be reading articles frequently this month. Qiita can be used more conveniently after logging in.
  1. We will deliver articles that match you
    By following users and tags, you can catch up information on technical fields that you are interested in as a whole
  2. you can read useful information later efficiently
    By "stocking" the articles you like, you can search right away
ユーザーは見つかりませんでした