Background removal models and runtime

U2NET and U2NETP by Xuebin Qin, Zichen Zhang, Chenyang Huang,
Masood Dehghan, Osmar R. Zaiane and Martin Jagersand.
U^2-Net: Going Deeper with Nested U-Structure for Salient Object Detection.
Pattern Recognition 2020. https://github.com/xuebinqin/U-2-Net
Apache License 2.0: LICENSE-U2NET.txt in this directory.

Unmodified ONNX model exports distributed by Daniel Gatis / rembg:
https://github.com/danielgatis/rembg/releases/tag/v0.0.0
u2net.onnx and u2netp.onnx verified against checksums in rembg sessions.
rembg MIT license: LICENSE-rembg.txt in this directory.
The site's inference and brush UI is separately implemented.

ONNX Runtime Web 1.30.0, Copyright (c) Microsoft Corporation.
Distributed unchanged from the onnxruntime-web npm package (MIT).
https://github.com/microsoft/onnxruntime
Runtime license and third-party notices:
/vendor/onnxruntime-1.30.0/LICENSE.txt
/vendor/onnxruntime-1.30.0/ThirdPartyNotices.txt

The model/runtime files are part of the hosted tool. Test photos, build
packages, benchmarks, development notes and screenshots belong in _dev
and are not required for production hosting.