aflw2k3d

• Description:

AFLW2000-3D is a dataset of 2000 images that have been annotated with image-level 68-point 3D facial landmarks. This dataset is typically used for evaluation of 3D facial landmark detection models. The head poses are very diverse and often hard to be detected by a cnn-based face detector. The 2D landmarks are skipped in this dataset, since some of the data are not consistent to 21 points, as the original paper mentioned.

• Additional Documentation: Explore on Papers With Code north_east

• Homepage: http://www.cbsr.ia.ac.cn/users/xiangyuzhu/projects/3DDFA/main.htm

• Source code: tfds.datasets.aflw2k3d.Builder

• Versions:

  • 1.0.0 (default): No release notes.

• Download size: 83.36 MiB

• Dataset size: 42.48 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	2,000

• Feature structure:

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),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
image	Image	(450,	uint8
: : : 450, : : :
: : : 3) : : :
landmarks_68_3d_xy_normalized	Tensor	(68,	float32
: : : 2) : : :
landmarks_68_3d_z	Tensor	(68,	float32
: : : 1) : : :

• Supervised keys (See as_supervised doc): None

• Figure (tfds.show_examples):

• Examples (tfds.as_dataframe):

{% framebox %}

Display examples...

const data = await response.text();
contentPane.innerHTML = data;

} catch (e) { contentPane.innerHTML = 'Error loading examples. If the error persist, please open ' + 'a new issue.'; } });

{% endframebox %}

• Citation:

@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}
}