scene_parse150

• Description:

Scene parsing is to segment and parse an image into different image regions associated with semantic categories, such as sky, road, person, and bed. MIT Scene Parsing Benchmark (SceneParse150) provides a standard training and evaluation platform for the algorithms of scene parsing.

• Additional Documentation: Explore on Papers With Code

• Homepage: http://sceneparsing.csail.mit.edu/

• Source code: tfds.datasets.scene_parse150.Builder

• Versions:

  • 1.0.0 (default): No release notes.

• Download size: 936.97 MiB

• Dataset size: 904.91 MiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'test'	2,000
'train'	20,210

• Feature structure:

FeaturesDict({
    'annotation': Image(shape=(None, None, 3), dtype=uint8),
    'image': Image(shape=(None, None, 3), dtype=uint8),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
annotation	Image	(None, None, 3)	uint8
image	Image	(None, None, 3)	uint8

• Supervised keys (See as_supervised doc): ('image', 'annotation')

• Figure (tfds.show_examples): Not supported.

• Examples (tfds.as_dataframe):

• Citation:

@inproceedings{zhou2017scene,
title={Scene Parsing through ADE20K Dataset},
author={Zhou, Bolei and Zhao, Hang and Puig, Xavier and Fidler, Sanja and Barriuso, Adela and Torralba, Antonio},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
year={2017}
}