# AI-Powered Masking

Fater offers a powerful way to **generate mask suggestions using AI based on text descriptions**. Instead of manually drawing complex selections, you can prompt the AI to identify objects or regions, and it will create a new layer representing that selection as a mask. You can then apply this generated mask layer to your main working mask.

***

### How "Prompt to Mask" Works

1. **Your Input:** You provide a text prompt describing the object or region you want to create a mask for (e.g., "the red car," "the sky," "all windows").
2. **AI Processing:**
   * Fater sends your prompt and the current visual context of your **Inpaint Canvas** (specifically, the content within the **Generation Area Bounding Box**) to a specialized AI segmentation model.
   * The AI model analyzes the image and attempts to identify the pixels corresponding to your text description.
3. **New Mask Layer Created:** The AI returns a white on transparent image where white represents the selected area transparent represents the unselected area. Fater takes this image and:
   * Adds it as a **new layer** to your project, often named something like "mask: \[your prompt]".
   * This new mask layer will appear as a thumbnail in the **Generation Task List** (just like other AI generations).
   * The result is **automatically added to your Layers Tab**, set to be invisible by default.
4. **Applying the Generated Mask Layer:** This new layer white surface will be automatically added or subtracted (if you Alt clicked the button) to your current mask.

***

### Using the "Prompt to Mask" Tool

1. **Prepare Your Canvas:** Ensure the image content you want to segment is visible on the canvas and within the Generation Area.
2. **Enter Your Segmentation Prompt:**
   * In the **Floating Prompt Area**, type a concise description of what you want to select.
   * *Examples:* `the main building`, `all trees`, `the product label`.
3. **Initiate Prompt to Mask:**
   * On the **Inpaint Toolbar**, click the **`TextSearch`** icon (🔍).
   * This action signals you want to generate a mask from a prompt. You'll then use the main **Floating Prompt Area** to enter your description.
4. **Refine:** Your main mask is now updated. You can refine it further using manual tools if needed.
5. **Restore**: Using the wand button on the mask layer will restore the mask if needed later.

***

**Tips for Effective AI Mask Generation:**

* **Be Specific but Concise:** Clear, unambiguous prompts work best.
* **Generation Area Matters:** The AI only analyzes the content within the current Generation Area Bounding Box.
* **Use as a Starting Point:** AI-generated masks are powerful starting points. You can often get complex selections quickly and then refine them manually.
* **Combine with Manual Tools:** Don't hesitate to add to or subtract from an AI-generated mask (once applied to the main mask) using the Brush, Lasso, or Rectangle tools.
* **Re-use Mask Layers:** Since the AI generates an actual layer, you can keep these "mask layers" in your project (perhaps in a separate Layer Workspace) and re-apply them later if needed.

This AI-powered mask generation creates a flexible workflow, providing intelligent selections as image layers that you can then integrate into your main mask.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://fater.gitbook.io/fater-ai-docs/using-editing-and-masking-tools/ai-powered-masking.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.