https://github.com/robertbrada/PyQt-image-annotation-tool

Tool for assigning labels to images from a given folder.
https://github.com/robertbrada/PyQt-image-annotation-tool

Last synced: 7 days ago
JSON representation

Tool for assigning labels to images from a given folder.

• Host: GitHub
• URL: https://github.com/robertbrada/PyQt-image-annotation-tool
• Owner: robertbrada
• Created: 2020-02-01T09:32:44.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2024-07-10T09:19:29.000Z (4 months ago)
• Last Synced: 2024-08-02T17:35:26.926Z (3 months ago)
• Language: Python
• Homepage:
• Size: 25.4 KB
• Stars: 102
• Watchers: 5
• Forks: 38
• Open Issues: 8
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# PyQt Image Annotation Tool

This app is used to label images in a given directory.

Labeled images can be moved or copied into sub-directories, which are named as assigned labels.

The app is just a single Python script with GUI.

![PyQt Image Annotation Tool GUI](https://i.stack.imgur.com/iihhf.png)

## What can this app do

For example you have folder ./data/images/ with a lot of images and you need to assign some

label(s) to these images.

- it can assign multiple labels to one image

- it allows you to choose number and names of your labels

- it can move/copy images to folders that are named as desired labels.

- it can generate .csv file with assigned labels.

- it can generate .xlsx file with assigned labels.

- all settings are handled via GUI

## Installation and usage

1. Clone the project:

    ```bash

    git clone https://github.com/robertbrada/PyQt-image-annotation-tool.git

    ```

2. Enter the directory and install the dependencies (you might need to use ```pip3``` instead of ```pip```):

    ```bash

    cd PyQt-image-annotation-tool

    pip install -r requirements.txt

    ```

3. Run the app (use ```python3``` for Python 3)

   ```bash

    python main.py

    ```

## Keyboard shortcuts

- Right Arrow : Next image

- Left Arrow : Previous image

- 1-9: Select label

## Contributing

Pull requests are welcomed.