wkentaro /labelme
Image annotation with Python. Supports polygon, rectangle, circle, line, point, and AI-assisted annotation.
★ 16,321⑂ 3,729👁 147⑃ 1,729● 1,013▤ 115
+4
Daily
16,321
stars
3,729
forks
147
watchers
1,729
prs
1,013
issues
115
releases
README
labelme
Image annotation with Python.
--
Installation
| Usage
| Examples
| labelme.io ↗
Community --
Youtube FAQ --
Description
Labelme is a graphical image annotation tool inspired by .\
It is written in Python and uses Qt for its graphical interface.
Looking for a simple install without Python or Qt? Get the standalone app at labelme.io.
\
VOC dataset example of instance segmentation.
\
Other examples (semantic segmentation, bbox detection, and classification).
\
Various primitives (polygon, rectangle, circle, line, and point).
Multi-language support (English, 中文, 日本語, 한국어, Deutsch, Français, and more).
Features
- [x] Image annotation for polygon, rectangle, circle, line and point (tutorial)
- [x] Image flag annotation for classification and cleaning (166)
- [x] Video annotation (video annotation)
- [x] GUI customization (predefined labels / flags, auto-saving, label validation, etc) (144)
- [x] Exporting VOC-format dataset for semantic segmentation, instance segmentation
- [x] Exporting COCO-format dataset for instance segmentation
- [x] AI-assisted point-to-polygon/mask annotation by SAM, EfficientSAM models
- [x] AI text-to-annotation by YOLO-world, SAM3 models
🌏 Available in 21 languages - English · 日本語 · 한국어 · 简体中文 · 繁體中文 · Deutsch · Ελληνικά · Français · Español · Italiano · Português · Nederlands · Magyar · Русский · ไทย · Tiếng Việt · Türkçe · Українська · Polski · Bahasa Indonesia · فارسی (LANG=ja_JP.UTF-8 labelme)
Installation
There are 3 options to install labelme:
Option 1: Using pip
For more detail, check "Install Labelme using Terminal"
bash
pip install labelme
To install the latest version from GitHub:
pip install git+https://github.com/wkentaro/labelme.git
Option 2: Using standalone executable (Easiest)
If you're willing to invest in the convenience of simple installation without any dependencies (Python, Qt),
you can download the standalone executable from "Install Labelme as App".
It's a one-time payment for lifetime access, and it helps us to maintain this project.
Option 3: Linux distribution packages
On some Linux distributions, labelme is also packaged in the system's native repository and can be installed with the distribution's standard package tooling. The badge below tracks which distributions currently ship labelme and which version each one provides:
Supported Python and platforms
| | Supported (v7.x) | Maintenance (v6.3.x) |
| ------ | ------------------------------ | -------------------- |
| Python | 3.12 - 3.14 | 3.10 - 3.11 |
| Qt | Qt6 (PySide6) | Qt5 |
| OS | 64-bit macOS / Windows / Linux | older OSes |
labelme follows SPEC 0 (the successor to NEP 29) for dropping Python versions, in step with its core scientific dependencies (numpy, scipy, scikit-image). v6.3.x is the maintenance line for Qt5 and Python 3.10 / 3.11 stragglers.
v6.3.x receives critical fixes only, on a best-effort basis with no release cadence or SLA. "CTop issues
#270Is it possible to display only specific classes?7 comments · 2018-12-06#833how to draw continulous points by tracking mouse when create polygon?[Feature]6 comments · 2021-02-01#876Auto-save feature not working5 comments · 2021-06-24#615--flags or --labels for text file displayed as name of text file instead of list in text file5 comments · 2020-03-04#1164i want to selected a box and move it,but always draw a point that's make so bad to use? it's very boring and make lower work efficency4 comments · 2022-08-26#1515Too many annotation points make the program prone to lagging. 3 comments · 2024-11-19#1213Mark lines for training but get areas3 comments · 2022-11-12#2541feat(ai): prompt-free Segment Everything mode2 comments · 2026-08-19
Discussions
#2947YOLOVAL: game-detection predictions as Labelme rectangles for human review (paid ONNX models)0 comments · 2026-11-04#2763Tiled rendering for large COG (geotiff) files, with an optional second image for before/after comparison2 comments · 2026-10-15#2724Suggested feature addition: Marking a circle with three points3 comments · 2026-09-29#2691Select the neighboring shape with Ctrl+Arrow keys (spatial navigation on the canvas)1 comments · 2026-09-15#2432Label in Annotation List turn white when selected2 comments · 2026-07-30#2291request for few feature0 comments · 2026-07-04#2290How to the predefined label list or remove a label from it2 comments · 2026-07-03#2267Suggestion: Make program work natively with YOLO style label (*.txt)1 comments · 2026-07-01
Latest release
v7.11.0 v7.11.0 2026-11-14Changed - SAM 3 asks to be downloaded again, as 1.8 GB on a GPU or 3.7 GB on the CPU, and asks again for its CPU files when AI falls back to the CPU. Its files from earlier versions stay on disk. (3099) - AI no longer retries a request on the CPU when the GPU fails partway through; the next request loads the model again. The status bar no longer shows "Preparing image…". (3100) Fixed - The model menu shows the download size for the AI device: SAM 3 shows about 1.8 GB on a GPU instead of its CPU size. (3104) - Deleting an AI model no longer freezes the window while AI runs, and other loaded