What is InPaint?

No matter how good your prompt and model are, it is rare to get a perfect image in one shot.

Inpainting is an indispensable way to fix small defects or add small changes to your generated images. 

In this post, we will go through a few basic examples to use inpainting for fixing defects.

If you are new to AI images, you may want to read the step-by-step guide first. 

Basic inpainting settings

After you upload an image to ControlImage section, you will be able to check one of the control options: ControlImage or InPaint. Please note that selecting both at the same time is not possible and one should be unchecked in order to select the other one.

The next thing we need to do is to mask the area of the image where we want to apply changes.

Click the Confirm Button.

After just a few seconds, a tooltip will appear in the upper right corner, confirming the mask selection.

We can then add the necessary changes for the prompt, in the Positive Prompt field.

- In this example, we used - modern house extension - prompt.

After that, we can click on the Generate button again.

Below are some of the generated results for our InPainted model.

Inpainting is an iterative process. You can apply it as many times as you want to refine an image.

Tips for inpainting

Successful inpainting requires patience and skill. Here are some take homes for using inpainting

• One small area at a time.

• Keep masked content at Original and adjust the sampling method, sampling steps and CFG Scale to get better results.

• Play with masked content to see which one works the best.

• If nothing works well within NovArch’s settings, use photo editing software like Photoshop or GIMP to paint the area of interest with the rough shape and color you wanted. Upload that image and inpaint with original content.