tucan9389 / SemanticSegmentation-CoreML

The example project of inferencing Semantic Segementation using Core ML
https://github.com/motlabs/awesome-ml-demos-with-ios
MIT License
320 stars 32 forks source link

Segment whole image #6

Closed wuharvey closed 3 years ago

wuharvey commented 3 years ago

Hi there - first off, great work on this repo! :D

I wonder if there's a way to segment the whole image by padding the sides - as it stands, since imageCropAndScaleOption is .centerCrop, we only get the center.

tucan9389 commented 3 years ago

There are three options:

And I think the .scaleFill option is that you want.

image source: https://developer.apple.com/documentation/vision/vncoremlrequest/2890144-imagecropandscaleoption

tucan9389 commented 3 years ago

@wuharvey If you want to ask more, reopen the issue feel free.