Convert labelme annotations into coco format in one step
Project description
labelme2coco Python Package for Linux/MacOS/Windows
Make your own dataset for object detection/instance segmentation using labelme and transform the format to coco json format.
Convert LabelMe annotations to COCO format in one step
labelme is a widely used is a graphical image annotation tool that supports classification, segmentation, instance segmentation and object detection formats. However, widely used frameworks/models such as Yolact/Solo, Detectron, MMDetection etc. requires COCO formatted annotations.
You can use this package to convert labelme annotations to COCO format.
Getting started
Installation
pip install -U labelme2coco
Basic Usage
labelme2coco path/to/labelme/dir
labelme2coco path/to/labelme/dir --train_split_rate 0.85
Advanced Usage
# import package
import labelme2coco
# set directory that contains labelme annotations and image files
labelme_folder = "tests/data/labelme_annot"
# set export dir
export_dir = "tests/data/"
# set train split rate
train_split_rate = 0.85
# convert labelme annotations to coco
labelme2coco.convert(labelme_folder, export_dir, train_split_rate)
# import functions
from labelme2coco import get_coco_from_labelme_folder, save_json
# set labelme training data directory
labelme_train_folder = "tests/data/labelme_annot"
# set labelme validation data directory
labelme_val_folder = "tests/data/labelme_annot"
# set path for coco json to be saved
export_dir = "tests/data/"
# create train coco object
train_coco = get_coco_from_labelme_folder(labelme_train_folder)
# export train coco json
save_json(train_coco.json, export_dir+"train.json")
# create val coco object
val_coco = get_coco_from_labelme_folder(labelme_val_folder, coco_category_list=train_coco.json_categories)
# export val coco json
save_json(val_coco.json, export_dir+"val.json")
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
labelme2coco-0.2.1.tar.gz
(17.4 kB
view hashes)
Built Distribution
Close
Hashes for labelme2coco-0.2.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 186ea3628d7d4dd4e676acefec7a91778010e810744894a39fe9a3947f32c411 |
|
MD5 | f92f4ecfc45aad53739dc72427b76b7a |
|
BLAKE2b-256 | 6e7eb6222ad746e7b89f225efab4b42edfcafae0662ee23283caaf7bd3ca3ca6 |