skip to navigation
skip to content

labelme 2.7.0

Annotation Tool for Object Segmentation.

<img src="" align="right"/>

labelme: Image Annotation Tool with Python

[![PyPI Version](](
[![Travis Build Status](](
[![Docker Build Status](](

Labelme is a graphical image annotation tool inspired by <http:"">.
It is written in Python and uses Qt for its graphical interface.


- Ubuntu / macOS / Windows
- Python2 / Python3
- [PyQt4 / PyQt5](


There are options:

- Platform agonistic installation: Anaconda, Docker
- Platform specific installation: Ubuntu, macOS


You need install [Anaconda]( then run below:

# python2
conda create --name=labelme python=2.7
source activate labelme
conda install pyqt
pip install labelme

# python3
conda create --name=labelme python=3.6
source activate labelme
# conda install pyqt
pip install pyqt5 # pyut5 can be installed via pip on python3
pip install labelme


You need install [docker](, then run below:

chmod u+x labelme_on_docker

# Maybe you need on macOS
./labelme_on_docker static/apc2016_obj3.jpg -O static/apc2016_obj3.json


sudo apt-get install python-qt4 pyqt4-dev-tools
sudo pip install labelme # python2 works


brew install pyqt # maybe pyqt5
pip install labelme # both python2/3 should work



Run `labelme --help` for detail.

labelme # Open GUI
labelme tutorial/apc2016_obj3.jpg # Specify file
labelme tutorial/apc2016_obj3.jpg -O tutorial/apc2016_obj3.json # Close window after the save

<img src=".readme/apc2016_obj3_screenshot.jpg" width="50%"/>

The annotations are saved as a [JSON]( file. The
file includes the image itself.


To view the json file quickly, you can use utility script:

labelme_draw_json tutorial/apc2016_obj3.json

<img src=".readme/apc2016_obj3_draw_json.jpg" width="70%"/>

**Convert to Dataset**

To convert the json to set of image and label, you can run following:

labelme_json_to_dataset tutorial/apc2016_obj3.json -o tutorial/apc2016_obj3_json

It generates standard files from the JSON file.

- [img.png](tutorial/apc2016_obj3_json/img.png): Image file.
- [label.png](tutorial/apc2016_obj3_json/label.png): Int32 label file.
- [label_viz.png](tutorial/apc2016_obj3_json/label_viz.png): Visualization of `label.png`.
- [label_names.txt](tutorial/apc2016_obj3_json/label_names.txt): Label names for values in `label.png`.

Note that loading `label.png` is a bit difficult
(`scipy.misc.imread`, `` may not work correctly),
and please use `` to avoid unexpected behavior:

# see tutorial/ also.
>>> import numpy as np
>>> import PIL.Image

>>> label_png = 'tutorial/apc2016_obj3_json/label.png'
>>> lbl = np.asarray(
>>> print(lbl.dtype)
>>> np.unique(lbl)
array([0, 1, 2, 3], dtype=int32)
>>> lbl.shape
(907, 1210)


<img src=".readme/screencast.gif" width="70%"/>


This repo is the fork of [mpitid/pylabelme](
whose development has already stopped.
File Type Py Version Uploaded on Size
labelme-2.7.0.tar.gz (md5) Source 2018-02-06 319KB