Saved datasets
Last updated
Download format
Usage rights
License from data provider
Please review the applicable license to make sure your contemplated use is permitted.
Cost to access
Described as free to access or have a license that allows redistribution.
8 datasets found
  1. portrait-segmentation

    Updated Aug 28, 2020
  2. r

    Portrait Img Segmentation

    Updated Apr 29, 2020
  3. - Matting Human Datasets

    Updated Jun 6, 2019
  4. o

    On Improving Face Generation for Privacy Preservation

    Updated Sep 4, 2019
  5. s

    Most interesting art investment products to collectors wolrdiwde2019

    Updated Nov 1, 2019
  6. w

    CGAP Smallholder Household Survey 2016, Building the Evidence Base on the...

    Updated Mar 12, 2019
  7. s

    Revenue of Samsung Electronics by business segment 2011-2019, by quarter

    Updated Nov 4, 2019
  8. w

    CGAP Smallholder Household Survey 2016, Building the Evidence Base on The...

    Updated Dec 13, 2017
  9. Not seeing a result you expected?
    Learn how you can add new datasets to our index.

Click to copy link
Link copied - Matting Human Datasets

Binary segmentation of humans and their background.

zip (15487364345 bytes)Available download formats
Dataset updated Jun 6, 2019
Laurent H.

Other (specified in description)



Human segmentation, i.e. high resolution extraction of humans from images, is a fascinating application with many uses. However, the problem is significantly under-constrained, making it an active area of research for developing more advanced methods. This dataset, developed by AISegment aims to help by providing a solid quality dataset of images and masks.

Quoting from the dataset author's GitHub (translated via Google Translate):

This dataset is currently the largest portrait matting dataset, containing 34,427 images and corresponding matting results. The data set was marked by the high quality of Beijing Play Star Convergence Technology Co., Ltd., and the portrait soft segmentation model trained using this data set has been commercialized.
The original images in the dataset are from Flickr, Baidu, and Taobao. After face detection and area cropping, a half-length portrait of 600*800 was generated.
The clip_img directory is a half-length portrait image in the format jpg; the matting directory is the corresponding matting file (convenient to confirm the matting quality), the format is png, you should first extract the alpha map from the png image before training. For example, using opencv you can get an alpha map like this:
In_image = cv2.imread('png image file path', cv2.IMREAD_UNCHANGED) Alpha = in_image[:,:,3]


See the author's GitHub.


This dataset comes in two parts:
1. Full images
2. The respective RGB "masks" or "cutouts" of those images


Thanks to the folks from SegmentAI for putting this dataset together.

Clear search
Close search
Google apps
Main menu