U2NetP model for local background removal

Original U^2-Net: Xuebin Qin, Zichen Zhang, Chenyang Huang, Masood Dehghan,
Osmar R. Zaiane, Martin Jagersand.
U^2-Net: Going Deeper with Nested U-Structure for Salient Object Detection (2020).
https://github.com/xuebinqin/U-2-Net
Apache License 2.0; full text in LICENSE.txt in this directory.

Unmodified ONNX model obtained from rembg by Daniel Gatis and contributors:
https://github.com/danielgatis/rembg/releases/download/v0.0.0/u2netp.onnx
Upstream verification: MD5 8e83ca70e441ab06c318d82300c84806
SHA-256: 309c8469258dda742793dce0ebea8e6dd393174f89934733ecc8b14c76f4ddd8

Only background segmentation is used; the site does not generate new photographs.
No affiliation or endorsement by the above authors is claimed.