Label Studio Annotation Tool

Label Studio is a open-source, web-based tool for annotating images, videos, audio files, and text. It allows users to create bounding boxes, polygons, points, and text annotations to efficiently label datasets for machine learning and AI applications. Label Studio is flexible, lightweight, and supports local deployment.

Features of Label Studio:

• Supports multiple annotation types including rectangles, polygons, points, and text classification.
• Runs in a web browser and can be self-hosted locally.
• Export annotations in formats like JSON, CSV, COCO, YOLO, and TFRecord.
• Supports image, video, audio, and text annotation.
• Provides customizable labeling interfaces.

Label Studio Version Check and Usage

$ sudo su
$ mkdir -p ~/labelstudio
$ cd ~/labelstudio
$ sudo docker compose up -d
$ sudo docker ps
$ sudo docker logs labelstudio -f
$ sudo docker exec -it labelstudio python -c "import label_studio; print(label_studio.__version__)"

Access Label Studio Web Interface

Open your browser and go to:
http://SERVER-IP:8080