Skip to content

Latest commit

 

History

History
31 lines (26 loc) · 1.18 KB

README.md

File metadata and controls

31 lines (26 loc) · 1.18 KB
Raw

JupyterImageAnnotator

A simple example of image data annotation in Jupyter Notebook with Jupyter Widgets. This isn't ready-to-use library. This is a template that you can easily customize for your annotation tasks. Advantages of data annotation with Jupyter Widgets:

  • ease of data manipulation
  • flexible customization of existing logic
  • fast experiments
  • it's just another Notebook in your project!

For example, it's very easy to add a button with a listener function:

bad_button = widgets.Button(description='Bad', button_style='danger')
bad_button.on_click(on_button_click)

Or add a widget with an image:

image_widget = widgets.Image(width=400, height=400)
with open(image_path, 'rb') as f:
    image_widget.value = f.read()

Example project - car annotation

Main requirements:

  • annotating the images of cars in two classes: "good" and "bad"(details in the notebook)
  • ability to save car ids of different classes into separate files
  • ability to continue annotating from the last image after reloading the notebook

All this requirements can be fullfilled in an hour.

The final form interface:

Feel free to use this template in your projects!