Visio ML is a Mac tool for annotating and augmenting image sets for use in object-detection machine-learning model training.
An Apple Mac computer running macOS 10.15 or later.
Open the project using Xcode and hit run (Command+R).
- Select a folder containing source images.
- Annotate each image by selecting it on the navigator and dragging areas over the preview.
- Augment your data set by appliying filters and generating annotated synthetic images.
- Export annotations as JSON file for use with Create ML.
- Open folder on Create ML to train your model.
Using Xcode 11 or later.
The project is written in Swift 5.2 using the SwiftUI and Combine frameworks.
Contact us if you're interested in contributing. Submit your PRs via GitHub for consideration.