/DepthPrediction-CoreML

The example of running Depth Prediction using Core ML

Primary LanguageSwiftMIT LicenseMIT

DepthPrediction-CoreML

platform-ios swift-version lisence

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

GIF demo 1 Screenshot 1 Screenshot 2 Screenshot 3 Screenshot 4

How it works

When use Metal

image

Requirements

  • Xcode 10.2+
  • iOS 11.0+
  • Swift 5

Model

Download

Download model from apple's model page.

Matadata

input node output node size
FCRN [1, 304, 228, 3]
name: image		 [1, 128, 160]
name: depthmap		 254.7 MB
FCRNFP16 [1, 304, 228, 3]
name: image		 [1, 128, 160]
name: depthmap		 127.3 MB

Inference Time

Device Inference Time Total Time(GPU) Total Time(CPU)
iPhone 12 Pro Max ⏲ ⏲ ⏲
iPhone 12 Pro ⏲ ⏲ ⏲
iPhone 12 ⏲ ⏲ ⏲
iPhone 12 Mini ⏲ ⏲ ⏲
iPhone 11 Pro Max ⏲ ⏲ ⏲
iPhone 11 Pro 134 ms 134 ms 149 ms
iPhone 11 ⏲ ⏲ ⏲
iPhone SE(2nd) ⏲ ⏲ ⏲
iPhone XS Max 146 ms ⏲ 155 ms
iPhone XS 146 ms ⏲ 151 ms
iPhone XR 148 ms ⏲ 154 ms
iPhone X 624 ms ⏲ 640 ms
iPhone 8+ 621 ms ⏲ 634 ms
iPhone 8 626 ms ⏲ 639 ms
iPhone 7+ 595 ms ⏲ 609 ms
iPhone 7 612 ms ⏲ 624 ms
iPhone 6S+ 1038 ms ⏲ 1051 ms

See also