googlesamples / googlesamples/mlkit
Pose detection keypoints are little dislocated in iOS
- Dominant language
- Kotlin
- Stars
- 4.3k
- Forks
- 3.1k
- PR merge metrics
- No merged PRs in 30d
Description
I've tried MLKit sample app (ios-> quickstarts ->vision). It seems pose keypoints are dislocated on iOS.
Please look into [sample](https://drive.google.com/file/d/186czWlzrypaUWGVRKCFQwzY_bfLVi33q/view?usp=sharing).
if we look into this sample, face key points (eyes, nose, ears, etc.. ) are totally dislocated.
I guess but not exactly, points are not properly converted with the following function:
> private func convertedPoints(
> from points: [NSValue]?,
> width: CGFloat,
> height: CGFloat
> ) -> [NSValue]? {
> return points?.map {
> let cgPointValue = $0.cgPointValue
> let normalizedPoint = CGPoint(x: cgPointValue.x / width, y: cgPointValue.y / height)
> let cgPoint = previewLayer.layerPointConverted(fromCaptureDevicePoint: normalizedPoint)
> let value = NSValue(cgPoint: cgPoint)
> return value
> }
> }
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.