Image Segmentation

Image Segmentation allows developers to partition a video or image into multiple segments that represent everyday things. As an example, image segmentation can help identify the outline of people walking in the street or discern the shapes of everyday things in your living room like couches and chairs.

Custom Training Models for Image Segmentation

You can train a custom model that is compatible with the Image Segmentation API by using Quickstart: Use Fritz AI Studio to Train a Custom Model.

Pre-trained Models

Include our models directly in your app and use them with the API.

Name Example Description Classes Mask Resolution
People people_seg A model that detects people masks person 384x384
People Medium people_seg A model that detects people masks. This model is slower then the one above but creates a higher quality mask. person 768x768
Pet pet_seg A model that detects pet masks. pet 224x224
Hair hair_seg A model that detects hair masks. hair 224x224
Sky sky_seg A model that detects sky masks. sky 224x224
Outdoor outdoor_seg A model that detects object masks in an outdoor scene. Building, Sky, Tree, Sidewalk, Ground, Car, Water, House, Fence, Fencing, Signboard, Sign, Skyscraper, Bridge, Span, River, Bus, Truck, Van, Motorbike, Bicycle, Traffic Light, Person 384x384
Living Room living_room_seg A model that detects object masks in a living room scene. Chair, Wall, Coffee Table, Ceiling, Floor, Bed, Lamp, Sofa, Window, Pillow 384x384

Technical Specifications

Architecture Format(s) Size Input Output Benchmarks
MobileNet and ICNet variants Core ML (iOS), TensorFlow Lite (Android) ~25 MB 224x224-pixel image Height and width of the mask, Number of classes the model predicts, Probability that pixel belongs to class 30 FPS on iPhone X, 10 FPS on Pixel 2