five

soltustiks/cyclegan-urban-vangogh

收藏
Hugging Face2026-04-20 更新2026-04-26 收录
下载链接:
https://hf-mirror.com/datasets/soltustiks/cyclegan-urban-vangogh
下载链接
链接失效反馈
官方服务:
资源简介:
--- license: mit task_categories: - image-to-image language: - en tags: - cyclegan - style-transfer - van-gogh - unpaired-translation pretty_name: Urban ↔ Van Gogh CycleGAN Dataset size_categories: - 1K<n<10K --- # Urban ↔ Van Gogh Dataset Unpaired image dataset for training CycleGAN to translate between photographs of European cities and paintings by Vincent van Gogh. ## Structure ​``` urban2vangogh/ ├── trainA/ (828 images) — Urban photographs (Unsplash) ├── trainB/ (350 images) — Van Gogh paintings ├── testA/ (80 images) └── testB/ (50 images) ​``` ## Domain A: Urban Photography Collected via Unsplash API. Categories: - European historic cities: Prague, Paris, Amsterdam, Vienna, Italy, Portugal, Spain - Urban nature: parks, boulevards, riversides - General urban landscapes ## Domain B: Van Gogh Paintings Source: [Berkeley CycleGAN Datasets](https://efrosgans.eecs.berkeley.edu/cyclegan/datasets/) (vangogh2photo, trainA split — 400 paintings total, with train/test split). ## Usage ​```python from huggingface_hub import hf_hub_download zip_path = hf_hub_download( repo_id="soltustiks/urban2vangogh", filename="urban2vangogh_dataset.zip", repo_type="dataset" ) ​``` ## License MIT for the collected subset. Van Gogh paintings are in the public domain. ## Citation Based on CycleGAN (Zhu et al. 2017).
提供机构:
soltustiks
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作