/ImageSegmentation-CoreML

The example of running Object Segmentation using Core ML

Primary LanguageSwiftMIT LicenseMIT

ImageSegmentation-CoreML

platform-ios swift-version lisence

This project is Object Segmentation on iOS with Core ML.
If you are interested in iOS + Machine Learning, visit here you can see various DEMOs.

Screenshot 1 Screenshot 2 Screenshot 3 Screenshot 4

How it works

(Preparing...)

Requirements

  • Xcode 10.2+
  • iOS 12.0+
  • Swift 5

Model

Download

Download model from apple's model page.

Matadata

input node output node size
DeepLabV3 [1, 513, 513, 3]
name: image
[513, 513]
name: semanticPredictions
8.6 MB
DeepLabV3FP16 [1, 513, 513, 3]
name: image
[513, 513]
name: semanticPredictions
4.3 MB
DeepLabV3Int8LUT [1, 513, 513, 3]
name: image
[513, 513]
name: semanticPredictions
2.3 MB

Inference Time

Device Inference Time Total Time
iPhone XS Max 133 ms 434 ms
iPhone XS 135 ms 411 ms
iPhone XR 133 ms 402 ms
iPhone X 178 ms 509 ms
iPhone 8+ 180 ms 563 ms
iPhone 8 189 ms 529 ms
iPhone 7+ 240 ms 667 ms
iPhone 7 247 ms 688 ms
iPhone 6S + 309 ms 1015 ms
iPhone 6+ 1888 ms 2753 ms

See also