A simple PyTorch Implementation of Generative Adversarial Networks, focusing on anime face drawing.

Last update: Jan 03, 2023

Overview

AnimeGAN

A simple PyTorch Implementation of Generative Adversarial Networks, focusing on anime face drawing.

Randomly Generated Images

The images are generated from a DCGAN model trained on 143,000 anime character faces for 100 epochs.

Image Interpolation

Manipulating latent codes, enables the transition from images in the first row to the last row.

Original Images

The images are not clean, some outliers can be observed, which degrades the quality of the generated images.

Usage

To run the experiment,

$ python main.py --dataRoot path_to_dataset/

The pretrained model for DCGAN are also in this repo, play it inside the jupyter notebook.

anime-faces Dataset

Anime-style images of 126 tags are collected from danbooru.donmai.us using the crawler tool gallery-dl. The images are then processed by a anime face detector python-animeface. The resulting dataset contains ~143,000 anime faces. Note that some of the tags may no longer meaningful after cropping, i.e. the cropped face images under 'uniform' tag may not contain visible parts of uniforms.

How to construct the dataset from scratch ?

Prequisites: gallery-dl, python-animeface

Download anime-style images

# download 1000 images under the tag "misaka_mikoto"
gallery-dl --images 1000 "https://danbooru.donmai.us/posts?tags=misaka_mikoto"

# in a multi-processing manner
cat tags.txt | \
xargs -n 1 -P 12 -I 'tag' \ 
bash -c ' gallery-dl --images 1000 "https://danbooru.donmai.us/posts?tags=$tag" '

Extract faces from the downloaded images

import animeface
from PIL import Image

im = Image.open('images/anime_image_misaka_mikoto.png')
faces = animeface.detect(im)
x,y,w,h = faces[0].face.pos
im = im.crop((x,y,x+w,y+h))
im.show() # display

I've cleaned the original dataset, the new version of the dataset has 115085 images in 126 tags. You can access the images from:

Brine (a python-based dataset management library): https://www.brine.io/jayleicn/anime-faces
Google Drive: https://drive.google.com/file/d/0B4wZXrs0DHMHMEl1ODVpMjRTWEk/view?usp=sharing
BaiduYun: https://pan.baidu.com/s/1o8Nxllo

Non-commercial use please.

Things I've learned

GANs are really hard to train.
DCGAN generally works well, simply add fully-connected layers causes problems.
In my cases, more layers for G yields better images, in the sense that G should be more powerful than D.
Add noise to D's inputs and labels helps stablize training.
Use differnet input and generate resolution (64x64 vs 96x96), there seems no obvious difference during training, the generated images are also very similar.
Binray Noise as G's input amazingly works, but the images are not as good as those with Gussian Noise, idea credit to @cwhy ['Binary Noise' here I mean a sequence of {-1,1} generated by bernoulli distribution at p=0.5 ]

I did not carefully verify them, if you are looking for some general GAN tips, see @soumith's ganhacks

Others

This project is heavily influenced by chainer-DCGAN and IllustrationGAN, the codes are mostly borrowed from PyTorch DCGAN example, thanks the authors for the clean codes.
Dependencies: pytorch, torchvision
This is a toy project for me to learn PyTorch and GANs, most importantly, for fun! :) Any feedback is welcome.

@jayleicn

Comments

$KeyError: 'module name can\'t contain$

KeyError: 'module name can\'t contain "."'

The classes in module.py contains some nn.Module layers whose names contains some'.' in it, so I got error messages like the title, so how could I play it??

opened by bolin12 2
no image under some tags

Hi, The dataset from google drive contains 126 tags. However, some folders are emtpy:

1girl apron blush collarbone hairclip honma_meiko japanese_clothes monochrome necktie nishizumi_miho purple_eyes scarf school_uniform sunglasses

Is this normal? Thanks

opened by samrere 1
set_sizes_contiguous is not allowed on a Tensor created from .data or .detach().

input.data.resize_(real_cpu.size()).copy_(real_cpu) RuntimeError: set_sizes_contiguous is not allowed on a Tensor created from .data or .detach(). If your intent is to change the metadata of a Tensor (such as sizes / strides / storage / storage_offset) without autograd tracking the change, remove the .data / .detach() call and wrap the change in a with torch.no_grad(): block. For example, change: x.data.set_(y) to: with torch.no_grad(): x.set_(y)

opened by athulvingt 0
can not run

Traceback (most recent call last): File "main.py", line 6, in import torch File "/Library/Python/2.7/site-packages/torch/init.py", line 81, in from torch._C import * RuntimeError: module compiled against API version 0xa but this version of numpy is 0x9

opened by xtzero 0
How do I start with my own model

I would like to know how I can use this image generation to generate my own images from a self made model?

Where can I read upon on this. I find no concrete info on making the models.

opened by quintendewilde 0

A simple PyTorch Implementation of Generative Adversarial Networks, focusing on anime face drawing.

Related tags

Overview

AnimeGAN

Randomly Generated Images

Image Interpolation

Original Images

Usage

anime-faces Dataset

Things I've learned

Others

Comments

KeyError: 'module name can\'t contain "."'

no image under some tags

set_sizes_contiguous is not allowed on a Tensor created from .data or .detach().

can not run

How do I start with my own model

Releases(data)

data(Aug 24, 2022)

Owner

Jie Lei 雷杰

Code and data for ImageCoDe, a contextual vison-and-language benchmark

ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation

Code release to accompany paper "Geometry-Aware Gradient Algorithms for Neural Architecture Search."

Introducing neural networks to predict stock prices

A Topic Modeling toolbox

Object recognition using Azure Custom Vision AI and Azure Functions

Gradient Inversion with Generative Image Prior

BOOKSUM: A Collection of Datasets for Long-form Narrative Summarization

PyTorch implementation of the TTC algorithm

Codes for our IJCAI21 paper: Dialogue Discourse-Aware Graph Model and Data Augmentation for Meeting Summarization

Deep Reinforcement Learning for Multiplayer Online Battle Arena

Code reproduce for paper "Vehicle Re-identification with Viewpoint-aware Metric Learning"

Tensorflow 2 implementations of the C-SimCLR and C-BYOL self-supervised visual representation methods from "Compressive Visual Representations" (NeurIPS 2021)

Tensorflow implementation of Character-Aware Neural Language Models.

Pipeline for employing a Lightweight deep learning models for LOW-power systems

Pixel Consensus Voting for Panoptic Segmentation (CVPR 2020)

Scaling Vision with Sparse Mixture of Experts

Companion code for the paper "An Infinite-Feature Extension for Bayesian ReLU Nets That Fixes Their Asymptotic Overconfidence" (NeurIPS 2021)

PyTorch implementation for our AAAI 2022 Paper "Graph-wise Common Latent Factor Extraction for Unsupervised Graph Representation Learning"

This is the formal code implementation of the CVPR 2022 paper 'Federated Class Incremental Learning'.