Fritz AI is the machine learning platform for iOS and Android developers. Teach your mobile apps to see, hear, sense, and think. Start with our ready-to-use feature APIs or connect and deploy your own custom models.
Vision API: Prebuilt models that you can simply drop into your apps:
- Image Segmentation: Create pixel level masks of different objects in a scene (code)
- Image Labeling: Classify different objects in an video or image (code)
- Pose Estimation: Identify and track a person's body position (code)
- Object Detection: Detect multiple objects and track their location (code)
- Style Transfer: Transform photos and videos into artistic masterpieces (code)
Custom Models: Deploy, Monitor, and Update your own models:
We currently support both TensorFlow Lite and TensorFlow Mobile for Android.
- Analytics and Monitoring: Monitor machine learning models running on-device with Fritz AI
- Model Management: Iterate on your ML models over-the-air, without having to release your app
- Model Protection: Use model protection to keep models from being tampered-with or stolen
- Android Studio 3.2 or above
- Android device in developer model (USB debugging enabled)
Step 1: Choose a Fritz AI Plan
Sign up for a Fritz AI plan to get started.
Register the Android app in your Fritz account with the package id "ai.fritz.aistudiobuild". During registration, you'll receive an API key for the app. Save this for later. To find it in the webapp, you can go to Project Settings > Apps > Your App > Show API Key.
Step 2: Clone / Fork the fritz-examples repository and open the demo app in Android Studio
git clone https://github.com/fritzlabs/fritz-examples.git
In Android Studio, choose "Open an existing Android Studio project" and select FritzAIStudio.
Step 3: Edit the fritz.xml file with your API Key
In app/src/main/res/values/fritz.xml, change the fritz_api_key attribute with the one you received in step 1.
Step 4: Build the Android Studio Project
Select "Build > Make Project" from the top nav. Download any missing libraries if applicable. This should sync the gradle dependencies so give the build a second to complete.
Step 5: Install the app onto your device
With your Android device connected, select Run > Run App from the top nav. After it's installed, select any of the options to try out the different ML features. When running the app for the first time, you'll have to give permissions to access the camera.
Heartbeat is a community of developers interested in the intersection of mobile and machine learning. Chat with us in Slack and stay up to date on the latest mobile ML news with our Newsletter.
For any questions or issues, you can: