{"url":"/dataset/afhq","name":"AFHQ","full_name":"Animal Faces-HQ","description_markdown":"Animal FacesHQ (AFHQ) is a dataset of animal faces consisting of 15,000 high-quality images at 512 × 512 resolution. The dataset includes three domains of cat, dog, and wildlife, each providing 5000 images. By having multiple (three) domains and diverse images of various\r\nbreeds (≥ eight) per each domain, AFHQ sets a more challenging image-to-image translation problem. \r\nAll images are vertically and horizontally aligned to have the eyes at the center. The low-quality images were discarded by human effort.\r\n\r\nSource: [StarGAN v2: Diverse Image Synthesis for Multiple Domains](https://arxiv.org/abs/1912.01865)","description_withheld":null,"homepage":"https://github.com/clovaai/stargan-v2","introduced_date":"2019-12-04","introduced_date_note":null,"introduced_by":{"paper":"/paper/stargan-v2-diverse-image-synthesis-for","title":"StarGAN v2: Diverse Image Synthesis for Multiple Domains","first_author":"Yunjey Choi","url":null},"license":{"name":"CC BY-NC 4.0","url":"https://github.com/clovaai/stargan-v2/blob/master/LICENSE"},"modalities":[{"name":"Images","url":"/datasets/modality/images"}],"tasks":[{"name":"Image Generation","url":"/task/image-generation","datasets_with_task":"/datasets/task/image-generation"},{"name":"Image-to-Image Translation","url":"/task/image-to-image-translation","datasets_with_task":"/datasets/task/image-to-image-translation"},{"name":"Multimodal Unsupervised Image-To-Image Translation","url":"/task/multimodal-unsupervised-image-to-image","datasets_with_task":"/datasets/task/multimodal-unsupervised-image-to-image"}],"languages":[],"variants":["AFHQ","AFHQV2","AFHQ Dog","AFHQ Cat","AFHQ Wild","AFHQcat2dog"],"data_loaders":[{"repo":"https://github.com/clovaai/stargan-v2","url":"https://github.com/clovaai/stargan-v2","frameworks":["pytorch"]}],"num_papers_in_archive":327,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/image-generation-on-afhq-cat","task":"Image Generation","dataset_variant":"AFHQ Cat","rows":8,"metrics":["clean-KID","clean-FID","FID","MAE Signature","MAE log-signature","RMSE Signature","RMSE log-signature"],"first_row_in_archive_order":{"model":"Vision-aided GAN","paper":"/paper/ensembling-off-the-shelf-models-for-gan","metrics":{"FID":"2.44","clean-FID":"2.51 ± .02","clean-KID":"0.46 ± .03"},"code_links":[{"title":"nupurkmr9/vision-aided-gan","url":"https://github.com/nupurkmr9/vision-aided-gan"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/image-generation-on-afhqv2","task":"Image Generation","dataset_variant":"AFHQV2","rows":7,"metrics":["FID","EQ-T","EQ-R"],"first_row_in_archive_order":{"model":"Polarity-StyleGAN3","paper":"/paper/polarity-sampling-quality-and-diversity","metrics":{"FID":"3.95"},"code_links":[{"title":"AhmedImtiazPrio/magnet-polarity","url":"https://github.com/AhmedImtiazPrio/magnet-polarity"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/image-generation-on-afhq-dog","task":"Image Generation","dataset_variant":"AFHQ Dog","rows":6,"metrics":["FID","clean-FID","clean-KID","MAE Signature","MAE log-signature","RMSE Signature","RMSE log-signature"],"first_row_in_archive_order":{"model":"Projected GAN","paper":"/paper/projected-gans-converge-faster","metrics":{"FID":"4.52"},"code_links":[{"title":"autonomousvision/projected_gan","url":"https://github.com/autonomousvision/projected_gan"},{"title":"dome272/ProjectedGAN-pytorch","url":"https://github.com/dome272/ProjectedGAN-pytorch"},{"title":"tsubota-kouga/ProjectedGAN","url":"https://github.com/tsubota-kouga/ProjectedGAN"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/image-generation-on-afhq-wild","task":"Image Generation","dataset_variant":"AFHQ Wild","rows":5,"metrics":["clean-KID","clean-FID","FID","MAE Signature","MAE log-signature","RMSE Signature","RMSE log-signature"],"first_row_in_archive_order":{"model":"Vision-aided GAN","paper":"/paper/ensembling-off-the-shelf-models-for-gan","metrics":{"FID":"2.25","clean-FID":"2.35 ± .02","clean-KID":"0.38 ± .02"},"code_links":[{"title":"nupurkmr9/vision-aided-gan","url":"https://github.com/nupurkmr9/vision-aided-gan"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/multimodal-unsupervised-image-to-image-5","task":"Multimodal Unsupervised Image-To-Image Translation","dataset_variant":"AFHQ","rows":4,"metrics":["FID"],"first_row_in_archive_order":{"model":"StarGAN v2","paper":"/paper/stargan-v2-diverse-image-synthesis-for","metrics":{"FID":"16.2"},"code_links":[{"title":"clovaai/stargan-v2","url":"https://github.com/clovaai/stargan-v2"},{"title":"naver-ai/StyleMapGAN","url":"https://github.com/naver-ai/StyleMapGAN"},{"title":"mindslab-ai/hififace","url":"https://github.com/mindslab-ai/hififace"},{"title":"kunheek/style-aware-discriminator","url":"https://github.com/kunheek/style-aware-discriminator"},{"title":"taki0112/StarGAN_v2-Tensorflow","url":"https://github.com/taki0112/StarGAN_v2-Tensorflow"},{"title":"eps696/stargan2","url":"https://github.com/eps696/stargan2"},{"title":"KbeautyHair/BaselineModel","url":"https://github.com/KbeautyHair/BaselineModel"},{"title":"karlchahine/neural-cover-selection-for-image-steganography","url":"https://github.com/karlchahine/neural-cover-selection-for-image-steganography"},{"title":"UdonDa/StarGAN-v2-pytorch-nonofficial","url":"https://github.com/UdonDa/StarGAN-v2-pytorch-nonofficial"},{"title":"SUPERSHOPxyz/stylegan3-gradient","url":"https://github.com/SUPERSHOPxyz/stylegan3-gradient"},{"title":"threeracha/Chuibbo-Flask-Server","url":"https://github.com/threeracha/Chuibbo-Flask-Server"},{"title":"zzz2010/starganv2_paddle","url":"https://github.com/zzz2010/starganv2_paddle"},{"title":"2023-MindSpore-4/Code7","url":"https://github.com/2023-MindSpore-4/Code7/tree/main/StarGAN"},{"title":"sss20young/Chuibbo-Flask-Server","url":"https://github.com/sss20young/Chuibbo-Flask-Server"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/image-to-image-translation-on-afhq","task":"Image-to-Image Translation","dataset_variant":"AFHQ","rows":2,"metrics":["LPIPS","FID"],"first_row_in_archive_order":{"model":"StarGAN v2","paper":"/paper/stargan-v2-diverse-image-synthesis-for","metrics":{"FID":"24.4","LPIPS":"0.524"},"code_links":[{"title":"clovaai/stargan-v2","url":"https://github.com/clovaai/stargan-v2"},{"title":"naver-ai/StyleMapGAN","url":"https://github.com/naver-ai/StyleMapGAN"},{"title":"mindslab-ai/hififace","url":"https://github.com/mindslab-ai/hififace"},{"title":"kunheek/style-aware-discriminator","url":"https://github.com/kunheek/style-aware-discriminator"},{"title":"taki0112/StarGAN_v2-Tensorflow","url":"https://github.com/taki0112/StarGAN_v2-Tensorflow"},{"title":"eps696/stargan2","url":"https://github.com/eps696/stargan2"},{"title":"KbeautyHair/BaselineModel","url":"https://github.com/KbeautyHair/BaselineModel"},{"title":"karlchahine/neural-cover-selection-for-image-steganography","url":"https://github.com/karlchahine/neural-cover-selection-for-image-steganography"},{"title":"UdonDa/StarGAN-v2-pytorch-nonofficial","url":"https://github.com/UdonDa/StarGAN-v2-pytorch-nonofficial"},{"title":"SUPERSHOPxyz/stylegan3-gradient","url":"https://github.com/SUPERSHOPxyz/stylegan3-gradient"},{"title":"threeracha/Chuibbo-Flask-Server","url":"https://github.com/threeracha/Chuibbo-Flask-Server"},{"title":"zzz2010/starganv2_paddle","url":"https://github.com/zzz2010/starganv2_paddle"},{"title":"2023-MindSpore-4/Code7","url":"https://github.com/2023-MindSpore-4/Code7/tree/main/StarGAN"},{"title":"sss20young/Chuibbo-Flask-Server","url":"https://github.com/sss20young/Chuibbo-Flask-Server"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/clr-gan-improving-gans-stability-and-quality","title":"CLR-GAN: Improving GANs Stability and Quality via Consistent Latent Representation and Reconstruction","date":"2024-09-30","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/ddmi-domain-agnostic-latent-diffusion-models","title":"DDMI: Domain-Agnostic Latent Diffusion Models for Synthesizing High-Quality Implicit Neural Representations","date":"2024-01-23","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":21,"samples_ran":17,"samples_unverified":4,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/bellman-optimal-step-size-straightening-of","title":"Bellman Optimal Stepsize Straightening of Flow-Matching Models","date":"2023-12-27","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":14,"samples_ran":8,"samples_unverified":6,"pointer_only_for_licence":14,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/class-continuous-conditional-generative","title":"Class-Continuous Conditional Generative Neural Radiance Field","date":"2023-01-03","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/genie-higher-order-denoising-diffusion","title":"GENIE: Higher-Order Denoising Diffusion Solvers","date":"2022-10-11","rows_on_this_dataset":3,"code_links":1,"syntology":null},{"paper":"/paper/diffusion-gan-training-gans-with-diffusion","title":"Diffusion-GAN: Training GANs with Diffusion","date":"2022-06-05","rows_on_this_dataset":3,"code_links":4,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":18,"samples_ran":13,"samples_unverified":5,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/signature-and-log-signature-for-the-study-of","title":"Signature and Log-signature for the Study of Empirical Distributions Generated with GANs","date":"2022-03-07","rows_on_this_dataset":3,"code_links":1,"syntology":null},{"paper":"/paper/polarity-sampling-quality-and-diversity","title":"Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values","date":"2022-03-03","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/ensembling-off-the-shelf-models-for-gan","title":"Ensembling Off-the-shelf Models for GAN Training","date":"2021-12-16","rows_on_this_dataset":3,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":15,"samples_ran":9,"samples_unverified":6,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/projected-gans-converge-faster","title":"Projected GANs Converge Faster","date":"2021-11-01","rows_on_this_dataset":3,"code_links":3,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":49,"samples_ran":38,"samples_unverified":11,"pointer_only_for_licence":6,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/alias-free-generative-adversarial-networks","title":"Alias-Free Generative Adversarial Networks","date":"2021-06-23","rows_on_this_dataset":3,"code_links":7,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":31,"samples_ran":18,"samples_unverified":13,"pointer_only_for_licence":9,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/training-generative-adversarial-networks-with-2","title":"Training Generative Adversarial Networks with Limited Data","date":"2020-06-11","rows_on_this_dataset":3,"code_links":28,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":29,"samples_ran":5,"samples_unverified":24,"pointer_only_for_licence":4,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/stargan-v2-diverse-image-synthesis-for","title":"StarGAN v2: Diverse Image Synthesis for Multiple Domains","date":"2019-12-04","rows_on_this_dataset":2,"code_links":14,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":0,"samples_unverified":5,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/mode-seeking-generative-adversarial-networks","title":"Mode Seeking Generative Adversarial Networks for Diverse Image Synthesis","date":"2019-03-13","rows_on_this_dataset":1,"code_links":2,"syntology":null},{"paper":"/paper/diverse-image-to-image-translation-via","title":"Diverse Image-to-Image Translation via Disentangled Representations","date":"2018-08-02","rows_on_this_dataset":1,"code_links":7,"syntology":null},{"paper":"/paper/multimodal-unsupervised-image-to-image","title":"Multimodal Unsupervised Image-to-Image Translation","date":"2018-04-12","rows_on_this_dataset":1,"code_links":13,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":0,"samples_unverified":5,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":9,"samples_harvested":187,"samples_ran":108,"samples_unverified":79,"pointer_only_for_licence":33,"papers_with_no_sample_that_ran":2,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}