sam_annotation_tool_GUI - Decoder Only ONNX Version

Annotation GUI tailored for efficiently annotating large batches of images using the Segment Anything model from Meta. The GUI streamlines the annotation process, allowing users to annotate numerous images in a row seamlessly.

Installation

Before using this version of the SAM Annotation GUI, ensure that the raw images and their embeddings are stored in the input folder. The application expects to find the images and embeddings in this directory. Make sure the embeddings correspond to the raw images and are correctly named.

Setting up Virtual Environment and Installing Dependencies

1. Clone the repository:

   git clone https://github.com/Kirscher/sam_annotation_tool_GUI/

2. Navigate to the project directory:

   cd sam_annotation_tool_GUI

3. Create and activate a virutal environment:

   python -m venv venv
   source venv/bin/activate

4. Install dependencies:

   pip install -r requirements.txt

Usage

1. Run the GUI: In the venv, execute python sam_GUI.py to launch the GUI.
2. Annotate: Utilize the annotation tools provided to annotate images efficiently.
3. Save Annotations: Save the annotated images masks in compressed RLE format in json files in the output/ folder.

Commands

• Saving Segmentation Result:

  • Press the s key to save the segmentation result (if a mask has been generated).

• Mask Selection Mode:

  • Press the w key to use the model for prediction and enter the mask selection mode.
  • In the mask selection mode, you can press the a and d keys to switch between different masks.
  • Press the s key to save the segmentation result.
  • Press the w key to return to point selection mode. The model will predict based on this mask the next time.

• Iterative optimization of selected points:

  • Right-click on the areas you don't need and left-click on the areas you need but are not covered by the mask. A few points are enough.