HOW TO USE PHOTOSHOP’S GENERATIVE AI. FIREFLY IN PHOTOSHOP BY ADOBE

INSTALLING PHOTOSHOP BETA

Go to your Creative Cloud. Make sure you have a valid Creative Cloud account and sign in.

Select Check for Updates under Updates. Make sure to acquire the most recent version of this to ping the server.

After that, select beta apps. You’ll see an install similar to this, although I’ve already completed mine. Press install. When you click update here after previously installing the beta, Photoshop version 24.6.0 should be displayed. Ensure your subscription is up to date. This won’t function with CS6. I apologise.

We’re going to start with something reasonable before diving off the deep end.

PHOTO EXTENSION AND UNCROPPING
How many times have you taken a picture and then thought, “Hey, I wish I’d gone a little bit wider,” afterward?

Grab the crop tool. Let’s drag it out, make it a little bit bigger.

Make a selection with the rectangular marquee tool.

Press Generative Fill.

Hit Generate.

To complete this image, Adobe Sensei applies AI to make it appear as though I had taken a wider shot. Amazing stuff, really.

CHANGING OBJECT AND GENERATING CLOTHING

Another quick example.

With the Lasso tool, make a quick selection.

This time, select Generative Fill and fill up the field. Bingo and a red T-shirt.

CHANGING A BACKDROP
Change the background now. We’re going to use a Lasso tool and make a quick selection of the area surrounding that background. Just to demonstrate to you how effectively this works, I’m making a very arbitrary choice. You don’t want it to be too tight, in fact. You should give it some space.

Now, we’re going to hit Generate a fill. And why don’t we put in city? Hit Generate.

And we actually get three options in the Properties panel.

Each time we regenerate or change the prompt, it creates three more images. Choose the one you want to use.

CLOTHING CHANGE IN ADVANCE
This is the starting point I had in mind. Instead of giving her a skirt, let’s give her some jeans.

Use the lasso tool to make a selection. (Watch this advice to learn how crucial decisions are and what kind of selections to make for better outcomes using generative fill.)

Hit Generative Fill, type in jeans. Let’s look at variations.

We can either view the info panel’s thumbnails to see the variances. or utilise the toolbar’s arrow buttons. Check out this guide to blending variants.

These photos are entirely new and have never been seen before. Let me demonstrate. Why not convert this jacket into a plaid shirt instead?

Use the lasso tool to make a selection.

Type in plaid shirt. And look at that.

The outcomes you acquire from this process aren’t always flawless.

We could substitute another item here, like a tank top. This one appears to be rather accurate. Look at the arms and the shirt, for example. Take a look at how the area for the sleeves and other details were produced.

It has provided us with more hair. There are undoubtedly numerous problems to address and improvements can be addressed. These improvements will occur extremely quickly if the past performance of other generative models, such as Mid Journey, is any indication.

STREET EXTENSION
Let’s consider one more instance.

Grab our crop tool, and let’s expand this.

Grab the Rectangular Marquee tool now since we want to extend the street.

From the toolbar, select Generative Fill. Just hit Generate without filling out any of the prompts if you want to extend the document but not add anything.

.And look at that. That’s pretty amazing if you ask me.

WHO MAKES THE IMAGES AND WHERE?
So from where are these pictures coming? These pictures have never before been seen. Actually, the AI is producing them.

The AI is actually examining the image, analysing it, and determining what it is. Adobe didn’t begin with this last week, though. Actually, Adobe has been working on Adobe Sensei for ten years. As a result, it has already received some training on all the options and various activities that we have done with the sensei tools throughout the years. The generative component of Sensei, called Firefly, studies every image in Adobe Stock before using what it has learned to create new versions of these photos or portions of them. Therefore, it is producing brand-new photos. According to Adobe, a solution will be found to pay the contributors of Adobe stock. We’ll watch to see what they decide.