简体中文 | English
Introduction
LabelU is an open source data annotation tool that supports Chinese. At present, it has image annotation capabilities such as rectangle, polygon, point, line, classification, and caption. It can support detection, classification, segmentation, text transcription, Line detection, key point detection and other computer vision task scenarios. You can customize the annotation task by freely combining tools, and support COCO and MASK format data export.
Getting started
Install locally with miniconda
# Download and Install miniconda
# https://docs.conda.io/en/latest/miniconda.html
# Create virtual environment(python = 3.7)
conda create -n labelu python=3.7
# Activate virtual environment
conda activate labelu
# Install labelu
pip install labelu
# Start labelu, server: http://localhost:8000
labelu
Install for local development
# Download and Install miniconda
# https://docs.conda.io/en/latest/miniconda.html
# Create virtual environment(python = 3.7)
conda create -n labelu python=3.7
# Activate virtual environment
conda activate labelu
# Install peotry
# https://python-poetry.org/docs/#installing-with-the-official-installer
# Install all package dependencies
poetry install
# Start labelu, server: http://localhost:8000
uvicorn labelu.main:app --reload
# Update submodule
git submodule update --remote --merge
feature
- Uniform, Six image annotation tools are provided, which can be configured through simple visualization or Yaml
- Unlimited, Multiple tools can be freely combined to meet most image annotation requirements
- Universal, Support multiple data export formats, including LabelU, COCO, Mask
Scenes
Computer Vision
- Detection: Detection scenes for vehicles, license plates, pedestrians, faces, industrial parts, etc.
- Classification: Detection of object classification, target characteristics, right and wrong judgments, and other classification scenarios
- Semantic segmentation: Human body segmentation, panoramic segmentation, drivable area segmentation, vehicle segmentation, etc.
- Text transcription: Text detection and recognition of license plates, invoices, insurance policies, signs, etc.
- Contour detection: positioning line scenes such as human contour lines, lane lines, etc.
- Key point detection: positioning scenes such as human face key points, vehicle key points, road edge key points, etc.
Usage
Annotation Format
Communication
Welcome to the Opendatalab Wechat group!
Links
- labelU-Kit(Powered by labelU-Kit)
LICENSE
This project is released under the Apache 2.0 license.