Style Transfer

Object Detection In Action

Use Style Transfer to bring real-time artistic style transfer to your apps. Transform photos with beautiful patterns or turn them into masterpieces painted by history’s greatest artists. Models run in real-time so you can create great experiences.

Custom Training Models for Style Transfer

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

Pre-trained Models

Choose from 25 different pre-trained styles. These styles are trained on famous paintings and patterns.

Paintings
Style Input Output
Bicentennial Print from America: The Third Century by Roy Lichtenstein bicentennial_print_input bicentennial_print_output
Les Femmes d’Alger by Picasso femmes_input femmes_output
Head of a Clown head_of_clown_input head_of_clown_output
Horses on the Seashore by Giorgio de Chirico horses_on_the_seashore_input horses_on_the_seashore_output
The Poppy Field by Claude Monet poppy_field_input poppy_field_output
Ritmo Plastico by Gino Severini ritmo_plastico_input ritmo_plastico_output
Starry Night by Vincent Van Gogh starry_night_input starry_night_output
The Scream by Edvard Munch the_scream_input the_scream_output
The Trial by Sidney Nolan the_trial_input the_trial_output
Patterns
Style Input Output
Blue Arrow blue_arrow_input blue_arrow_output
Comic comic_input comic_output
Filament filament_input filament_output
Green Blocks green_blocks_input green_blocks_output
Kaleidoscope kaleidoscope_input kaleidoscope_output
Lamp Post lamp_post_input lamp_post_output
Mosaic mosaic_input mosaic_output
Notre Dame notre_dame_input notre_dame_output
Pink and Blue Rhombuses pink_blue_rhombus_input pink_blue_rhombus_output
Shades shades_input shades_output
Sketch sketch_input sketch_output
Snowflake snowflake_input snowflake_output
Sprinkles sprinkles_input sprinkles_output
Swirl swirl_input swirl_output
Tile tile_input tile_output
Vector vector_input vector_output

Technical Specifications

Architecture Format(s) Size Input Output Benchmarks
Fast Style Transfer Core ML (iOS), TensorFlow Lite (Android) 17 KB (8-bit quantization); 467 KB (stabilized) Arbitrary size images (iOS12); 640x480-pixel image Stylized image 28 FPS on iPhone X, 2 FPS on Pixel 2

Stable Style Transfer

For video, visually stabilized style transfer creates beautiful continuity between frames.