Skip to content
Semantic Segmentation iOS app that runs a model that is made using MakeML app.
Swift Objective-C++ Objective-C Ruby
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Pods
Segmentation Live.xcodeproj
Segmentation Live.xcworkspace
Segmentation Live
Podfile
Podfile.lock
README.md
chair_flower_segmentation.gif

README.md

Object Detection and Segmentation MakeML

MakeML is a Developer Tool for Creating Object Detection and Segmentation Neural Networks without a Line of Code. It's built to make the training process easy to setup. It is designed to handle data sets, training configurations, markup and training processes — all in one place.

Semantic Segmentation iOS

Is an iOS app example that uses Tensorflow Lite on GPU. Used DeepLab .tflite model.

Train Objects Segmentation .tflite model

MakeML object detection and segmentation ML models

See the Tutorial for the training object segmentation model without a line of code with macOS desktop application.

Using another .tflite model in iOS application

MakeML object detection and segmentation ML models MakeML object detection and segmentation ML models

For using this project with another .tflite file, add it to the project and change this line with your name of the model.

NSString *modelPath = FilePathForResourceName(@"deeplabv3_257_mv_gpu", @"tflite");

Change the number of classes and add colors to the array:

int class_count = 21;
unsigned int colors[21] = { ... };

Links

More Tutorials | MakeML in App Store | Full Documentation | MakeML Chat | Support page

You can’t perform that action at this time.