U2NET fashion image/clothing segmenter based on


The ClothingSegmenter executor can be used to perform segmentation on images of fashion products.
The executor is based on a U2NET architecture pre-trained on iMaterialistic fashion 2019. The
parts of the image that represent clothing or fashion items are recognized, using the U2NET pixel-wise
segmentation model and the rest of the image content is filtered out. Images that pass through
the executor are resized to a fixed shape of (500, 768) to match the pre-training image size.



via Docker image (recommended)

from jina import Flow
f = Flow().add(uses='jinahub+docker://ClothingSegmenter')

via source code

from jina import Flow
f = Flow().add(uses='jinahub://ClothingSegmenter')
  • To override __init__ args & kwargs, use .add(..., uses_with: {'key': 'value'})
  • To override class metas, use .add(..., uses_metas: {'key': 'value})


View Github