aflw2k3d

  • 説明:

AFLW2000-3D は、画像レベルの 68 ポイントの 3D 顔ランドマークで注釈が付けられた 2000 枚の画像のデータセットです。このデータセットは通常、3D 顔ランドマーク検出モデルの評価に使用されます。頭のポーズは非常に多様で、多くの場合、cnn ベースの顔検出器では検出が困難です。元の論文で言及されているように、一部のデータは 21 ポイントと一致していないため、このデータセットでは 2D ランドマークはスキップされます。

スプリット
'train' 2,000
  • 機能構造:
FeaturesDict({
    'image': Image(shape=(450, 450, 3), dtype=uint8),
    'landmarks_68_3d_xy_normalized': Tensor(shape=(68, 2), dtype=float32),
    'landmarks_68_3d_z': Tensor(shape=(68, 1), 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(450、450、3) uint8
ランドマーク_68_3d_xy_normalizedテンソル(68, 2) float32
ランドマーク_68_3d_zテンソル(68,1) float32

視覚化

  • 引用
@article{DBLP:journals/corr/ZhuLLSL15,
  author    = {Xiangyu Zhu and
               Zhen Lei and
               Xiaoming Liu and
               Hailin Shi and
               Stan Z. Li},
  title     = {Face Alignment Across Large Poses: {A} 3D Solution},
  journal   = {CoRR},
  volume    = {abs/1511.07212},
  year      = {2015},
  url       = {http://arxiv.org/abs/1511.07212},
  archivePrefix = {arXiv},
  eprint    = {1511.07212},
  timestamp = {Mon, 13 Aug 2018 16:48:23 +0200},
  biburl    = {https://dblp.org/rec/bib/journals/corr/ZhuLLSL15},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}