aflw2k3d

  • 설명 :

AFLW2000-3D는 이미지 수준 68포인트 3D 얼굴 랜드마크로 주석이 달린 2000개 이미지의 데이터 세트입니다. 이 데이터 세트는 일반적으로 3D 얼굴 랜드마크 감지 모델 평가에 사용됩니다. 머리 포즈는 매우 다양하며 종종 cnn 기반 얼굴 감지기로 감지하기 어렵습니다. 이 데이터 세트에서 2D 랜드마크는 건너뛰었습니다. 일부 데이터는 원래 논문에서 언급한 것처럼 21개 지점과 일치하지 않기 때문입니다.

나뉘다
'train' 2,000
  • 기능 구조 :
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),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
영상 영상 (450, 450, 3) uint8
Landmarks_68_3d_xy_normalized 텐서 (68, 2) float32
랜드마크_68_3d_z 텐서 (68, 1) float32

심상

  • 인용 :
@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}
}