enhancedlabelme 0.9

Image Polygonal Annotation with Python

EnhancedLabelMe

1. (Linux) > pip install enhancedlabelme (Windows : Run cmd as administrator) > python -m pip install enhancedlabelme 2. you can start the program typing below command line > labelme+ 3. (to update) > pip install enhancedlabelme --upgrade v0.1 (200805) - Added time keys to json data (loadTime, saveTime) - Removed double click function when creating polygon v0.2 (200806) - Always dump jsons in utf-8 encoding - "Save Automatically" is checked as default - Added indication of work progress in the window title (# of images/# of jsons) - Added the Preference in menu bar (File - Preference) and open config file. : You can change hotkey(단축키) in this config file. It will be applied after relaunch v0.3 (200807) - Labelme will launch in fullscreen mode as default v0.4 (200810) - Added the dark theme (menubar - View - Toggle dark theme) : To set dark theme as default, add below line in config file and relaunch > dark_mode: true dark_mode: true - Changed the icon to nuvi-labs v0.6 - Fixed bug when reading json fails due to encoding v0.7 - The enhanced labelme is separated from the labelme which means it will be installed as independent package. : you don't have to delete existing labelme from now on : Start enhanced labelme by typing below line in terminal > labelme+ - Fixed tool bar icon alignment v0.8 (200811) - Open config file with notepad.exe as default in Windows - Cleaned some useless codes v0.9 (200812) - Fixed unexpected error when closing file - Fixed writing wrong version to json(0.8.1 --> 4.1.1) - Added drag & drop function (drag & drop to open image except folder) - Added brightness & contrast adjustment function (View - Brightness Contrast) - Added RGB indicator right to labellist (same color with polygon) - Cleaned some codes

• To Do

• Add open recent directory menu bar
• drag & drop folder
• Enable deleting elements from label list
• Draw grid lines on image to help drawing bounding box

@misc{labelme2016, author = {Kentaro Wada}, title = {{labelme: Image Polygonal Annotation with Python}}, howpublished = {\url{https://github.com/wkentaro/labelme}}, year = {2016} }