# CycleGAN (ICCV'2017)
> [Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks](https://arxiv.org/abs/1703.10593)
<!-- [ALGORITHM] -->
## Abstract
<!-- [ABSTRACT] -->
Image-to-image translation is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image using a training set of aligned image pairs. However, for many tasks, paired training data will not be available. We present an approach for learning to translate an image from a source domain X to a target domain Y in the absence of paired examples. Our goal is to learn a mapping G:X→Y such that the distribution of images from G(X) is indistinguishable from the distribution Y using an adversarial loss. Because this mapping is highly under-constrained, we couple it with an inverse mapping F:Y→X and introduce a cycle consistency loss to push F(G(X))≈X (and vice versa). Qualitative results are presented on several tasks where paired training data does not exist, including collection style transfer, object transfiguration, season transfer, photo enhancement, etc. Quantitative comparisons against several prior methods demonstrate the superiority of our approach.
<!-- [IMAGE] -->
<div align=center >
<img src="https://user-images.githubusercontent.com/12726765/144200449-cc2777da-3112-4024-aaa6-c6be5c8220bc.png" width="400"/>
</div >
## Results and models
We use `FID` and `IS` metrics to evaluate the generation performance of CycleGAN.
| Method | FID | IS | Download |
| :---------------------------------------------------------------------------------------: | :---------: | :-------: | :-----------------------------------------------------------------------------------------: |
| official facades | 123.626 | **1.638** | - |
| [ours facades](/configs/synthesizers/cyclegan/cyclegan_lsgan_resnet_in_1x1_80k_facades.py) | **118.297** | 1.584 | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_facades/cyclegan_lsgan_resnet_in_1x1_80k_facades_20200524-0b877c2a.pth) \| [log](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_facades/cyclegan_lsgan_resnet_in_1x1_80k_facades_20200524_211816.log.json) |
| official facades-id0 | **119.726** | 1.697 | - |
| [ours facades-id0](/configs/synthesizers/cyclegan/cyclegan_lsgan_id0_resnet_in_1x1_80k_facades.py) | 126.316 | **1.957** | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_facades_id0/cyclegan_lsgan_id0_resnet_in_1x1_80k_facades_20200524-438aa074.pth) \| [log](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_facades_id0/cyclegan_lsgan_id0_resnet_in_1x1_80k_facades_20200524_212548.log.json) |
| official summer2winter | 77.342 | 2.762 | - |
| [ours summer2winter](/configs/synthesizers/cyclegan/cyclegan_lsgan_resnet_in_1x1_246200_summer2winter.py) | **76.959** | **2.768** | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_summer2winter/cyclegan_lsgan_resnet_in_1x1_246200_summer2winter_20200524-0baeaff6.pth) \| [log](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_summer2winter/cyclegan_lsgan_resnet_in_1x1_246200_summer2winter_20200524_214809.log.json) |
| official winter2summer | **72.631** | **3.293** | - |
| [ours winter2summer](/configs/synthesizers/cyclegan/cyclegan_lsgan_resnet_in_1x1_246200_summer2winter.py) | 72.803 | 3.069 | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_summer2winter/cyclegan_lsgan_resnet_in_1x1_246200_summer2winter_20200524-0baeaff6.pth) \| [log](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_summer2winter/cyclegan_lsgan_resnet_in_1x1_246200_summer2winter_20200524_214809.log.json) |
| official summer2winter-id0 | 76.773 | **2.750** | - |
| [ours summer2winter-id0](/configs/synthesizers/cyclegan/cyclegan_lsgan_id0_resnet_in_1x1_246200_summer2winter.py) | **76.018** | 2.735 | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_summer2winter_id0/cyclegan_lsgan_id0_resnet_in_1x1_246200_summer2winter_20200524-f280ecdd.pth) \| [log](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_summer2winter_id0/cyclegan_lsgan_id0_resnet_in_1x1_246200_summer2winter_20200524_215511.log.json) |
| official winter2summer-id0 | 74.239 | 3.110 | - |
| [ours winter2summer-id0](/configs/synthesizers/cyclegan/cyclegan_lsgan_id0_resnet_in_1x1_246200_summer2winter.py) | **73.498** | **3.130** | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_summer2winter_id0/cyclegan_lsgan_id0_resnet_in_1x1_246200_summer2winter_20200524-f280ecdd.pth) \| [log](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_summer2winter_id0/cyclegan_lsgan_id0_resnet_in_1x1_246200_summer2winter_20200524_215511.log.json) |
| official horse2zebra | **62.111** | 1.375 | - |
| [ours horse2zebra](/configs/synthesizers/cyclegan/cyclegan_lsgan_resnet_in_1x1_266800_horse2zebra.py) | 63.810 | **1.430** | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_horse2zebra/cyclegan_lsgan_resnet_in_1x1_266800_horse2zebra_20200524-1b3d5d3a.pth) \| [log](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_horse2zebra/cyclegan_lsgan_resnet_in_1x1_266800_horse2zebra_20200524_220040.log.json) |
| official horse2zebra-id0 | 77.202 | **1.584** | - |
| [ours horse2zebra-id0](/configs/synthesizers/cyclegan/cyclegan_lsgan_id0_resnet_in_1x1_266800_horse2zebra.py) | **71.675** | 1.542 | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_horse2zebra_id0/cyclegan_lsgan_id0_resnet_in_1x1_266800_horse2zebra_20200524-470fb8da.pth) \| [log](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_horse2zebra_id0/cyclegan_lsgan_id0_resnet_in_1x1_266800_horse2zebra_20200524_220655.log.json) |
| official horse2zebra | **138.646** | **3.186** | - |
| [ours zebra2horse](/configs/synthesizers/cyclegan/cyclegan_lsgan_resnet_in_1x1_266800_horse2zebra.py) | 139.279 | 3.093 | [model](https://download.openmmlab.com/mmediting/synthesizers/cyclegan/cyclegan_horse2zebra/cyclegan_lsgan_resnet_in_1x1_266800_horse2zebra_20200524-1b3d5d3a.pth) \| [log](https://download.openmmlab.com/mmediting/sy
没有合适的资源?快使用搜索试试~ 我知道了~
资源推荐
资源详情
资源评论
收起资源包目录
mmediting-master.zip (716个子文件)
make.bat 761B
make.bat 761B
CITATION.cff 319B
setup.cfg 606B
readthedocs.css 144B
readthedocs.css 144B
Dockerfile 819B
Dockerfile 527B
.gitignore 1KB
MANIFEST.in 226B
matting_tutorial.ipynb 604KB
restorer_basic_tutorial_zh-CN.ipynb 171KB
restorer_basic_tutorial.ipynb 170KB
zhihu_qrcode.jpg 388KB
mmediting-demo.jpg 227KB
qq_group_qrcode.jpg 200KB
GT26r.jpg 167KB
GT05.jpg 152KB
GT05.jpg 88KB
qq_group2_qrcode.jpg 70KB
4.jpg 63KB
3.jpg 58KB
5.jpg 58KB
2.jpg 53KB
3.jpg 52KB
2.jpg 51KB
1.jpg 47KB
1.jpg 43KB
6.jpg 40KB
GT05.jpg 18KB
test_list.json 413B
LICENSE 11KB
Makefile 634B
Makefile 634B
changelog.md 41KB
changelog.md 40KB
tools_scripts.md 16KB
tools_scripts.md 15KB
getting_started.md 11KB
getting_started.md 11KB
README.md 10KB
config_matting.md 9KB
config_generation.md 9KB
config_matting.md 9KB
README_zh-CN.md 8KB
config_restoration.md 8KB
config_generation.md 8KB
customize_models.md 8KB
config_restoration.md 8KB
config_inpainting.md 8KB
customize_dataset.md 8KB
config_inpainting.md 7KB
install.md 7KB
demo.md 7KB
demo.md 7KB
customize_pipeline.md 6KB
quick_run.md 6KB
quick_run.md 6KB
README.md 6KB
README_zh-CN.md 6KB
README.md 6KB
README.md 5KB
README.md 5KB
install.md 5KB
README.md 5KB
README.md 5KB
README.md 5KB
README_zh-CN.md 5KB
README.md 5KB
README_zh-CN.md 5KB
README.md 4KB
README.md 4KB
README.md 4KB
README.md 4KB
README_zh-CN.md 4KB
README.md 4KB
README.md 4KB
README.md 4KB
README.md 4KB
README.md 4KB
README.md 4KB
README.md 4KB
README.md 3KB
README.md 3KB
README.md 3KB
README.md 3KB
README.md 3KB
CODE_OF_CONDUCT.md 3KB
README.md 3KB
README_zh-CN.md 3KB
README.md 3KB
README_zh-CN.md 3KB
README_zh-CN.md 3KB
README_zh-CN.md 3KB
README_zh-CN.md 3KB
README.md 3KB
README_zh-CN.md 3KB
README.md 3KB
README.md 3KB
README_zh-CN.md 3KB
共 716 条
- 1
- 2
- 3
- 4
- 5
- 6
- 8
资源评论
m0_72731342
- 粉丝: 2
- 资源: 1832
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
最新资源
- python-leetcode面试题解之第198题打家劫舍-题解.zip
- python-leetcode面试题解之第191题位1的个数-题解.zip
- python-leetcode面试题解之第186题反转字符串中的单词II-题解.zip
- 一个基于python的web后端高性能开发框架,下载可用
- python-leetcode面试题解之第179题最大数-题解.zip
- python-leetcode面试题解之第170题两数之和III数据结构设计-题解.zip
- python-leetcode面试题解之第168题Excel表列名称-题解.zip
- python-leetcode面试题解之第167题两数之和II输入有序数组-题解.zip
- python-leetcode面试题解之第166题分数到小数-题解.zip
- python-leetcode面试题解之第165比较版本号-题解.zip
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功