New! Generate Entire AI Images in Photoshop with Adobe Firefly

Photoshop Essentials
6 May 202409:43

TLDRAdobe's Photoshop beta introduces 'Generate Image', a feature powered by Adobe Firefly that creates AI-generated images from text prompts. With capabilities up to 2048x2048 pixels, users can match styles with reference images and apply various effects. The feature is accessible through the Contextual Task Bar, toolbar icon, or Edit menu, offering a text-to-image module integrated directly into Photoshop. Users can experiment with different prompts, style references, and effects to craft unique images, enhancing the creative possibilities within Photoshop.

Takeaways

  • 🆕 Adobe has released new generative AI features in Photoshop beta called Generate Image, using Adobe Firefly.
  • 🎨 Generate Image can create AI images from scratch, with features similar to the Firefly website.
  • 🔍 It uses the Firefly Image 3 model for higher quality and resolution, up to 2048 x 2048 pixels.
  • 💡 Generate Image is currently available only in the Photoshop beta version.
  • 🖼️ To access it, Creative Cloud subscribers can find the Photoshop beta in the Creative Cloud Desktop app.
  • 📐 Users should create a new document with a resolution limit of 2048 pixels for width and height.
  • 🔎 The Generate Image option can be found in the Contextual Task Bar, toolbar, or Edit menu.
  • 📝 Users can enter a prompt to describe the desired image and choose a content type, either Art or Photo.
  • 🖼️ Reference images can be used to match colors and the overall look, and effects can be added to the image.
  • ⏱️ Generating an image takes about 10-15 seconds and provides three initial results to choose from.
  • 🔄 Users can regenerate images with adjusted prompts, reference images, or effects for different outcomes.

Q & A

  • What is the Adobe Firefly and how does it relate to Photoshop?

    -Adobe Firefly is a generative AI feature that Adobe has recently integrated into the Photoshop beta. It allows users to create AI images from scratch, utilizing text-to-image capabilities that were previously available on the Firefly website.

  • What is the new feature called that enables the creation of AI images in Photoshop?

    -The new feature is called 'Generate Image', and it is embedded directly into Photoshop, allowing users to generate entire AI images using Adobe Firefly.

  • How does the Generate Image feature leverage the Firefly Image 3 model?

    -The Generate Image feature uses the Firefly Image 3 model to enhance the quality and resolution of the generated images, allowing for higher resolution outputs up to 2048 x 2048 pixels.

  • What is the resolution limit for the images generated by the Firefly Image 3 model?

    -The resolution limit for images generated by the Firefly Image 3 model is 2048 x 2048 pixels.

  • How can one access the Generate Image feature in Photoshop beta?

    -To access the Generate Image feature, one needs to have a Creative Cloud subscription and install the Photoshop beta from the Creative Cloud Desktop app. It can be found under the 'Beta' section in the 'Apps' menu.

  • What is the process to create a new document in Photoshop for using the Generate Image feature?

    -To create a new document for the Generate Image feature, go to the 'File' menu and choose 'New', then set both the Width and Height to 2048 pixels, and click 'Create'.

  • Where can the Generate Image option be found within Photoshop?

    -The Generate Image option can be found in three places: in the Contextual Task Bar, at the bottom of the toolbar, and in the 'Edit' menu.

  • What is the purpose of entering a prompt in the Generate Image dialog box?

    -The prompt serves as a description of the image the user wants to create. It helps the AI understand the desired output and generate images accordingly.

  • How can a user change the style of the generated images?

    -Users can change the style of the generated images by using a reference image to match colors and visual style, or by adding different effects through the Effects section in the Generate Image dialog box.

  • What are the different content types available for generating images?

    -The two content types available for generating images are 'Art' for a more creative and artistic image, and 'Photo' for something more photorealistic.

  • How does the Generate Similar option work in the Generate Image feature?

    -The Generate Similar option allows users to select a preferred result, and then Photoshop uses that result as the base image to generate three more images that are similar in style and appearance.

Outlines

00:00

🎨 Introducing Adobe Photoshop's Generate Image Feature

Steve from Photoshop Essentials introduces a new feature in the Photoshop beta called Generate Image, which utilizes Adobe Firefly's text-to-image capabilities. This feature allows users to create high-resolution images (up to 2048 x 2048 pixels) from scratch based on text prompts. The tutorial covers how to access the beta version, set up a new document, and use the Generate Image dialog box. It explains the process of entering a prompt, selecting content type (Art or Photo), and generating initial image results. The video also demonstrates how to view and select from the initial results, and how to regenerate images with different prompts.

05:04

🖼️ Customizing AI-Generated Images with Style References and Effects

The second part of the tutorial focuses on customizing the style of AI-generated images by using reference images to match colors, lighting, and visual style. It guides viewers on how to edit the reference image and generate new results that align with the desired style. The video also explores the use of effects to further customize the images, with options divided into categories like Popular, Materials, and Techniques. Steve demonstrates adding multiple effects to achieve a 'papery' look and provides a step-by-step guide on combining prompts, reference images, and effects to generate unique images. The conclusion highlights the user's ability to regenerate images until satisfied with the results, showcasing the flexibility of the Generate Image feature in Photoshop beta.

Mindmap

Keywords

💡Adobe Firefly

Adobe Firefly is a generative AI technology developed by Adobe Inc. It is designed to create content such as images, based on textual descriptions or prompts. In the context of the video, Adobe Firefly is used to generate entire AI images directly within Photoshop, showcasing its integration with Adobe's suite of creative tools. The script mentions that it uses the 'Firefly Image 3 model' for higher quality and resolution, indicating its advanced capabilities.

💡Photoshop beta

Photoshop beta refers to a pre-release version of Adobe Photoshop, the widely used image editing software. Beta versions are typically released to a limited audience for testing and feedback before the official release. The script discusses the availability of the 'Generate Image' feature in the Photoshop beta, which allows users to experience the new capabilities before they are fully rolled out.

💡Generate Image

Generate Image is a feature within the Photoshop beta that leverages Adobe Firefly's AI to create images from scratch based on text prompts. It represents a significant advancement in AI-assisted image creation, offering users a new way to generate content. The script explains how to access and use this feature, including setting up a new document and entering prompts to create images.

💡Text-to-image module

The text-to-image module is a component of Adobe Firefly that converts textual descriptions into visual images. It is highlighted in the script as being embedded directly into Photoshop, allowing users to generate images by simply typing in a description. This module is central to the 'Generate Image' feature and represents a fusion of AI and creative software.

💡Reference image

A reference image in the context of the video is an existing image that users can select to guide the style, colors, and overall look and feel of the AI-generated image. The script explains how users can choose a reference image to influence the output of the 'Generate Image' feature, thereby customizing the results to match a specific aesthetic or mood.

💡Effects

Effects in the video script refer to additional visual styles or treatments that can be applied to the AI-generated images. They offer a way to further customize the look of the generated content. The script describes various categories of effects, such as 'Layered Paper', 'Newspaper', and 'Double Exposure', which can be selected to modify the final image.

💡Content Type

Content Type is a setting within the 'Generate Image' feature that allows users to specify the nature of the image they want to create. The script mentions two options: 'Art' for a more creative and artistic image, and 'Photo' for something more photorealistic. This setting helps guide the AI in generating images that align with the user's vision.

💡Resolution

Resolution in the context of the video pertains to the pixel dimensions of the images generated by the 'Generate Image' feature. The script specifies that the new Firefly Image 3 model can generate images up to 2048 x 2048 pixels, which is significantly higher than previous capabilities, allowing for larger and more detailed images.

💡Layers panel

The Layers panel in Photoshop is a user interface component that allows users to manage different layers of an image, each of which can represent different parts or effects of the composition. The script mentions the Layers panel in relation to the 'Generative layer', which holds the results of the AI-generated images, indicating how users can interact with and manage these AI-generated results.

💡Contextual Task Bar

The Contextual Task Bar in Photoshop is a customizable toolbar that provides quick access to tools and options relevant to the task at hand. The script describes how to access the 'Generate Image' feature through the Contextual Task Bar, emphasizing its convenience and the ease with which users can initiate the AI image generation process.

💡Creative Cloud subscription

A Creative Cloud subscription is a service provided by Adobe that gives users access to a suite of Adobe's creative software, including Photoshop. The script mentions that having a Creative Cloud subscription is a prerequisite for accessing the Photoshop beta, which includes the new 'Generate Image' feature, highlighting the value of the subscription for early access to innovative tools.

Highlights

Adobe has released new generative AI features in the Photoshop beta.

The Generate Image feature can create an entire AI image from scratch using Adobe Firefly.

Generate Image is embedded directly into Photoshop, bringing over features from the Firefly website.

It uses the new Firefly Image 3 model for better quality and higher resolution images.

Images can now be generated up to 2048 x 2048 pixels, four times higher than Generative Fill.

Generate Image is currently only available in the Photoshop beta.

Creative Cloud subscribers can access the Photoshop beta through the Creative Cloud Desktop app.

To use Generate Image, create a new document with a resolution limit of 2048 pixels.

The Generate Image option can be found in three places within Photoshop.

Enter a prompt to describe the image you want to create.

Choose a content type between Art for creative images or Photo for photorealistic ones.

Use a reference image to match colors and overall look and feel.

Add different effects to the image for additional styling.

Photoshop sends the file to Adobe Firefly servers to generate the image.

You receive three initial results to choose from and can generate similar images based on your selection.

Edit your prompt for different results or use a reference image to change the style.

Effects can be added or changed to achieve different visual styles.

Generate Image allows for a wide range of creative possibilities within Photoshop.