Object Selection

Create an editable image mask from a canvas prompt, then reuse it for background removal, adjustments, effects, extraction, or export.

Workflow

  1. Select an image and activate Select Object.
  2. Click an object to add a positive prompt. Hold Shift while clicking to subtract a region.
  3. Drag a box when a coarse region is easier to identify than a point.
  4. Add more prompts to refine the temporary preview. Press Escape to discard it.
  5. Choose Apply as mask when the preview is ready. Apply is one undoable document operation.

Models and privacy

The optional local model is downloaded explicitly from the AI Models settings, verified with its manifest checksum, and loaded lazily. Images are not uploaded by this workflow. Embeddings are temporary session data and are not stored in the document.

After applying, the document stores a normal raster mask resource, so reopening the document does not require the model to be installed.

Limitations

Object Selection is promptable segmentation, not semantic recognition. A click tells the model which region to segment; it does not guarantee that an arbitrary automatic result is the subject. Hair, fur, glass, smoke, shadows, and translucent edges may need brush or trimap refinement.