[barhanc]

Description

Adds generic foreground segmentation model for objects.

The other linked model (BiRefNet) is too heavy for on-device inference.

Introduces a breaking change?

• Yes
• No

Type of change

• Bug fix (change which fixes an issue)
• New feature (change which adds functionality)
• Documentation update (improves or adds clarity to existing documentation)
• Other (chores, tests, code style improvements etc.)

Tested on

• iOS
• Android

Testing instructions

• CV example app -> semantic segmentation screen -> run new model (xnnpack and coreml). Test on a non-person image and compare output with selfie segmentation.
• Check published HF repo https://huggingface.co/software-mansion/U2NetP-ExecuTorch/tree/main

Screenshots

Related issues

Closes #1569

Checklist

• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have updated the documentation accordingly
• My changes generate no new warnings

Additional notes

[msluszniak]

Ok, xnnpack and coreml worked correctly on iOS, now Android and we should be ready to go.

[msluszniak]

Android worked as well on both xnnpack and vulkan. :)