feiwu77777/animime-birds

Animime · Birds

15 North American bird species — curated from CUB-200 (cassiekang mirror, 100-class subset).

15 classes, organized as an ImageFolder tree under train/, val/, and test/ (80/10/10 deterministic split, seed=42).

Classes

• american_crow
• american_goldfinch
• anna_hummingbird
• belted_kingfisher
• cardinal
• cedar_waxwing
• common_raven
• dark_eyed_junco
• house_sparrow
• pileated_woodpecker
• purple_finch
• red_winged_blackbird
• roadrunner
• scarlet_tanager
• white_pelican

Source & attribution

Images derive from the Caltech-UCSD Birds 200-2011 (CUB-200-2011, via cassiekang/cub200_dataset 100-class HF mirror) dataset, filtered to the subset of classes above and re-split. Original source: https://www.vision.caltech.edu/datasets/cub_200_2011/

License: Research / non-commercial (CUB-200-2011 terms of use).

Citation:

Wah, C., Branson, S., Welinder, P., Perona, P., & Belongie, S. (2011). The Caltech-UCSD Birds-200-2011 Dataset. California Institute of Technology.

This repository redistributes a filtered subset of the original images in ImageFolder layout to make them easy to load for the animime classifier. All rights to the underlying images remain with the original dataset authors; if you use this data, please cite the source above.

Loading

from datasets import load_dataset
ds = load_dataset("feiwu77777/animime-birds")
print(ds["train"][0])

Or as a raw ImageFolder tree:

from torchvision.datasets import ImageFolder
train = ImageFolder("data/birds/train")

Reproducing

The filtering + splitting pipeline lives in scripts/download_datasets.py in the animime_model repo.