Sure. I’ll need an iOS engineer as well and god know what models are supported on device currently, so it’ll be 250k and I’ll need 6 months.

This is actually a reasonable estimate.

I mean it's not like there are no libraries for human segmentation which already run on these devices....

Just use detectron2

Sure you can probably get v0 working in a week or two (assuming you know mobile stuff) but ironing out weird edge cases and assumed features takes forever

Good luck with that without data coming from the "v0"