jakarto3d / jakarnotator

The Jakarnotator is an annotation tool to create your own database for instance segmentation problem.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Jakarnotator

The Jakarnotator is an annotation tool to create your own database for instance segmentation problem. Jakarnotator data are compatible with the format of MS Coco database.

example

The Jakarnotator is based on Leaflet. You can annotate any image format supported by Leaflet (.png, .jpeg, ...) and even georeferenced data. You just need to put those images into a specific folder. Getting started

At Jakarto, we used a Jakarnotator instance to create a database of urban furnitures. With these data we were able to train our models with the Mask RCNN algorithm (no modification needed to deal with input data format as Detectron (a Mask RCNN implementation) used MS Coco).

A live demo is available on jakarnotator.jakarto.com.

Links

To see the Jakarnotator in action please take a look to these two links :

Usage

Click here if you want to learn how to annotate objects in the Jakarnotator

Getting started

If you are even more interested in creating your database from your own images or create new classes of objects you should take a look to this GETTING_STARTED.md.

Contribute

Jakarnotator is open-sourced as a simple tool to ease the creation of instance segmentation databases. Feel free to open an issue for questions, details, new features...

About

The Jakarnotator is an annotation tool to create your own database for instance segmentation problem.


Languages

Language:JavaScript 67.7%Language:HTML 27.8%Language:CSS 4.3%Language:Dockerfile 0.2%