Segmentation Annotator

This repository contains a segmentation annotator tool that allows you to annotate images for segmentation tasks. The tool was used to annotate The Great Outdoors Dataset.

annotator_v2.1.mp4

Features

• Super Pixelation
• Auto "SAM" Masks
• Interactive SAM masks
• Rectangle SAM mask
• Coarse OFFseg segmentation
• Manual Polygon

Installation

1. Clone the repository:

   git clone https://github.com/kasiv008/Segmentation-annotator.git

2. Install the required dependencies:

   pip install -r requirements.txt

3. Install SAM package

   cd segment-anything; pip install -e .

Download the SAM checkpoint and update the path in gui.py Line:50.

Usage

1. Run the annotator tool:

   python gui.py

2. Load an image and start annotating.

3. Save your annotations and export the annotated image.

Custom Dataset

To annotate your custom dataset with your own custom labels, please refer to custom_annotations_readme.md

Contributing

Contributions are welcome! If you find any issues or have suggestions for improvements, please open an issue or submit a pull request.

We will release a new update with the integration of SAM 2 for faster video annotations. Stay tuned!

License

This project is licensed under the MIT License.