Smart Segmentation

Smart Segmentation allows you to divide your images into multiple segments and create pixel-accurate annotations fast.

Run Smart Segmentation

To run Smart Segmentation on an image:

  1. In Data, select one or multiple images. Note that the image resolution shouldn't exceed 4 MP.
  2. Click Run Smart Segmentation.
  3. Choose a model: Generic or Autonomous driving.
  4. Click Start segmentation.

When Smart Segmentation completes successfully, you'll see a green checkmark on the Smart Segmentation button that corresponds to the image(s). When you hover over that button, you'll see this message: Smart Segmentation completed.

When Smart Segmentation is running on an image, the image is inactive. When you hover over it you'll see this message: The item is currently unavailable.

Create an instance with Smart Segmentation

smart segmentation

To create an instance with Smart Segmentation:

  1. Run Smart Segmentation on an image. Learn more.
  2. Open the image in the editor.
  3. Under Number of segments in the left panel, select a number. These numbers determine how much you want to segment the image. The higher the number you choose, the more the image will be segmented, and vice versa.
  4. Click on the segment(s) that form the object you want to annotate.
  5. Press the spacebar on the keyboard to create the instance.