stanford_dogs

  • 説明:

Stanford Dogs データセットには、世界中の 120 種類の犬の画像が含まれています。このデータセットは、画像を細かく分類するタスクのために、ImageNet の画像と注釈を使用して構築されています。 20,580 枚の画像があり、そのうち 12,000 枚がトレーニングに使用され、8580 枚がテストに使用されます。クラス ラベルとバウンディング ボックスの注釈は、12,000 枚の画像すべてに提供されます。

スプリット
'test' 8,580
'train' 12,000
  • 機能構造:
FeaturesDict({
    'image': Image(shape=(None, None, 3), dtype=uint8),
    'image/filename': Text(shape=(), dtype=string),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=120),
    'objects': Sequence({
        'bbox': BBoxFeature(shape=(4,), dtype=float32),
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
画像画像(なし、なし、3) uint8
画像/ファイル名文章ストリング
ラベルクラスラベルint64
オブジェクト順序
オブジェクト/bbox BBoxFeature (4) float32

視覚化

  • 引用
@inproceedings{KhoslaYaoJayadevaprakashFeiFei_FGVC2011,
author = "Aditya Khosla and Nityananda Jayadevaprakash and Bangpeng Yao and
          Li Fei-Fei",
title = "Novel Dataset for Fine-Grained Image Categorization",
booktitle = "First Workshop on Fine-Grained Visual Categorization,
             IEEE Conference on Computer Vision and Pattern Recognition",
year = "2011",
month = "June",
address = "Colorado Springs, CO",
}
@inproceedings{imagenet_cvpr09,
        AUTHOR = {Deng, J. and Dong, W. and Socher, R. and Li, L.-J. and
                  Li, K. and Fei-Fei, L.},
        TITLE = { {ImageNet: A Large-Scale Hierarchical Image Database} },
        BOOKTITLE = {CVPR09},
        YEAR = {2009},
        BIBSOURCE = "http://www.image-net.org/papers/imagenet_cvpr09.bib"}