AI Background Remover: third-party model and runtime

U2-Net compact (u2netp), Xuebin Qin and contributors.
Original project: https://github.com/xuebinqin/U-2-Net
Model and project licence: Apache License 2.0, included as LICENSE.txt.
Original pretrained compact model is linked from the author's project.
Unmodified ONNX conversion distributed by rembg:
https://github.com/danielgatis/rembg/releases/download/v0.0.0/u2netp.onnx
SHA-256: 309c8469258dda742793dce0ebea8e6dd393174f89934733ecc8b14c76f4ddd8
MD5 (upstream checksum): 8e83ca70e441ab06c318d82300c84806

Normalisation and mask extraction follow the rembg reference implementation,
adapted for browser ImageData and JavaScript. rembg is copyright 2020-present
Daniel Gatis, MIT License (included as rembg-LICENSE.txt).
https://github.com/danielgatis/rembg

ONNX Runtime Web 1.22.0, copyright Microsoft Corporation, MIT License.
Runtime files are unmodified and hosted by this website, not loaded from a CDN.
Licence: /assets/js/vendor/onnxruntime-web-1.22.0/LICENSE.txt
Third-party notices: /assets/js/vendor/onnxruntime-web-1.22.0/ThirdPartyNotices.txt
https://github.com/microsoft/onnxruntime/tree/v1.22.0

This tool is independent of Adobe and is not affiliated with or endorsed by Adobe.