U2NetP background-removal model (unmodified ONNX artifact)

Upstream project: U-2-Net by Xuebin Qin and contributors
Source: https://github.com/xuebinqin/U-2-Net
Model/project license: Apache-2.0; see LICENSE.txt in this directory.
Release distribution: https://github.com/danielgatis/rembg/releases/download/v0.0.0/u2netp.onnx
Distribution project: https://github.com/danielgatis/rembg (MIT)
File size: 4,574,861 bytes
SHA-256: 309c8469258dda742793dce0ebea8e6dd393174f89934733ecc8b14c76f4ddd8

Image Batch uses this lightweight 320 x 320 foreground segmentation model locally.
The model weights are not modified. Predictions are not guaranteed to be correct.
It is not a dedicated hair/glass matting model. Review results before use.