https://github.com/bchao1/anime-generation
🎨 Anime generation with GANs.
https://github.com/bchao1/anime-generation
acgan anime anime-generation computer-vision conditional-gan dataset dcgan gan generative-adversarial-network
Last synced: 6 months ago
JSON representation
🎨 Anime generation with GANs.
- Host: GitHub
- URL: https://github.com/bchao1/anime-generation
- Owner: bchao1
- Created: 2018-09-25T03:51:39.000Z (about 7 years ago)
- Default Branch: master
- Last Pushed: 2022-05-12T06:03:35.000Z (over 3 years ago)
- Last Synced: 2025-03-26T08:37:06.905Z (7 months ago)
- Topics: acgan, anime, anime-generation, computer-vision, conditional-gan, dataset, dcgan, gan, generative-adversarial-network
- Language: Python
- Homepage:
- Size: 15.1 MB
- Stars: 69
- Watchers: 2
- Forks: 7
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# ACGAN Conditional Anime Generation
> Note: I have restructured the whole folder, cleaned up training code, pruned my dataset, and updated most of the results. You can see the old version of this repo in `old`.
Generate colorful anime characters using GAN.
## Generating characters with different hair and eye colors
By interpolating generator input, we can see some interesting results.
By fixing color classes and varying noise, we can generate anime characters with same colors but different identity.
## Generate characters with different artistic styles
I noticed that anime produced in different years have distinctive artistic styles. To name a few:
- Characters in older anime tend to have larger, angular eyes, while newer anime have characters with round eyes.
- Colors are brighter and more saturated in older anime, while newer anime have mild colors.
By conditioning the ACGAN on **year labels**, we can generate characters with different artistic styles.
Interpolating along the year latent code:
## Start training
Modify `config.yaml` as you wish.
```
> python3 run.py
```
## The dataset
The current [dataset](https://drive.google.com/file/d/18FG-KSeAUBp0O0np63ynqdA7_tgIOCa8/view?usp=sharing) is a composition of 2 datasets:
1. One dataset with eye and hair color labels (30k+ images)
2. One dataset with year label, which is the year in which the anime is produced (60k+ images).
The dataset format is as follows:
```
- images/
- XXXXX.jpg
- ...
- labels.pkl
- eye_label.json
- year_label.json
- hair_label.json
```
After loading in `labels.pkl` with pickle, you will get a dictionary of `{ filname : labels }`. The labels are formatted as `(eye, hair, year)` tuples.
```
{
"32455": (8, 10, 5),
...
}
```
> This means `32455.jpg` has eye class 8, hair class 10, year class 5.
Missing labels will be a `None`. All images from dataset 1 will have year labels `None`, while all images from dataset 2 will have eye and hair label `None`.
***
Source code in the current repo is used to train on the first dataset. This requires some manual preprocessing (see `dataset/anime_dataset.py`) to extract the first dataset from the whole dataset.
***
The `.json` files map discrete labels to semantics.
```js
// eye_label.json
{
"aqua": 0,
"black": 1,
...
}
```
## Some notes
- When training on the first dataset, adding some color transformations to images as a preprocessing step traning might help. You can achieve this through various `torchvisions.transforms.functional` methods.
- Train with N(0, 1) but sample from Gaussian of smaller variance when evaluating. Just an engineering hack to get better results.