AutoLabel is a labeling assistance feature that uses predictions from an ML model to automatically annotate an image. The model is pre-trained on the COCO dataset, a large dataset of common objects. AutoLabel can be used with the Rectangle and Polygon label types.
Read more about the large dataset used in the pre-trained model and the object categories it supports.
To use AutoLabel, you first have to configure the classes you want to use for each label. There are two ways to do this:
In the Label Definitions tab, click Add Label.
Enter details such as label name and color. Label type must be Rectangle or Polygon.
In the Classes (optional) drop-down, select one or more pre-trained classes from the list.
Click Save to add your label.
In the Labeler, select the Rectangle or Polygon label you want to use for AutoLabel predictions
Click the to select AutoLabel from the drop-down.
If classes were not configured in Label Definitions, you will be prompted to attach one or more classes to this label
Click "Okay" to apply the classes. AutoLabel will begin analyzing the image for predictions from the ML model based on the classes selected. Labels will be applied to any matches.
The AutoLabel icon will have a dark background when it's enabled for that label. To turn off Auto Label, simply click this icon to toggle it off.
AutoLabel works best when:
used for detecting common, everyday objects in images
used as assistance for labeling images quickly. Additional labeling or adjusting of the labels by a human labeler may be necessary