Custom keypoint detection model through vision api

Question

Created Sep ’25

Replies 1

Boosts 0

Views 1k

Participants 2

Hi there,

I have a custom keypoint detection model and want to use it via vision's CoremlRequest API. Here's some complication for input and output:

For input My model expect 512x512 a image. Which would be resized and padded from a 1920x1080 frame. I use the .scaleToFit option, but can I also specify the color used for padding?

For output:

My model output a CoreMLFeatureValueObservation, can I have it output in a format vision recognizes? such as joints/keypoints
If my model is able to output in a format vision recognizes, would it take care to restoring the coordinates back to the original frame? (undo the padding) If not, how do I restore it from .scaletofit option?

Best,

Answered by olokobayusuf in 861420022

Were you able to figure this out? You will probably have to implement the coordinate transformations manually.

Answer 1

olokobayusuf OP

Oct ’25

Accepted Answer

Were you able to figure this out? You will probably have to implement the coordinate transformations manually.