Random code snippets while learning CoreML - an iOS framework for running trained Machine Learning models.
- Real-time capture using AVCaptureSession.
- Uses a CoreML Model (Resnet50.mlmodel) to detect objects within the captured data.
- Reports the detected objects and confidence level.
- User can choose a photo from their Photo Library.
- The Vision Framework is used to detect faces in the chosen photo.
- The detected bounding box around each face is drawn as a red rectangle in the image. Face bounding box remain when the photo is rotated from portrait to landscape.
- Introducing CoreML - WWDC 2017 presentation
- CoreML In Depth - WWDC 2017 presentation
- CoreML Tools - for converting trained Machine Learning models for use in iOS apps.
- Twitter: @clintcabanero
- GitHub: ccabanero