The News: Adobe announced the beta availability of its Adobe Firefly Image 3 Foundation Model, which is designed to provide greater quality and control to users, including higher quality image generations, better prompt understanding, more detail and variety to images, and other improvements to support faster ideation and creation of images. In addition, Adobe unveiled the all-new Photoshop, which is now in beta. It includes advancements in Generative Fill with Reference Image and is powered by Firefly Image 3, delivering greater control and power for creators. You can read about the Firefly Image 3 foundation model here and learn about the enhancements to Photoshop at this link.
Adobe Announces Firefly Image 3 Model and Photoshop Enhancements
Analyst Take: At Adobe MAX London, Adobe announced the beta availability of Firefly Image 3 Foundation Model and a new beta version of Photoshop with advanced capabilities. The enhancements are designed to bring a new level of quality and detail to AI-generated content, as well as provide more functionality within Photoshop.
Firefly Image 3 Foundation Model Delivers Advanced Quality and Capabilities
Adobe has released its latest generative AI foundation model for Firefly, called Image 3, which is designed to provide users with greater power and functionality when generating images, either within the Firefly web application or within Photoshop. The Firefly Image 3 is now able to deliver a number of new features, including:
- Auto-stylization capabilities: Auto-stylization capabilities in Firefly Image 3 deliver higher quality outputs with more variety, giving users more control and personalization over the styles of images they generate. Image outputs include new varieties of styles, colors, backgrounds, subject poses, and more.
- Advanced creative control with Structure Reference and Style Reference: Firefly Image 3-powered Structure Reference and Style Reference capabilities provide more user control and state-of-the-art visual quality. Structure Reference enables users to quickly generate new images that match the structure of a reference, whereas Style Reference provides additional control and personalization over the styles they generate. Combined, these capabilities let users reference both the structure and style of an image to quickly bring their ideas to life without needing to create a perfect prompt, saving time and effort.
- Increased power: Firefly Image 3 increased power is designed to deliver new levels of photographic quality with better lighting, positioning, and more variety. Significant improvements can be seen in the rendering of people and objects, such as the ability to render straight lines, adding realism into generated imagery.
- Greater prompt understanding: Firefly Image 3 has a greater capability to understand text prompts and scenes, enabling image generations that accurately reflect long, complex prompts and include richer details. Text renderings are also improved, resulting in image generations with clear text displays, helping to ensure that images with text reflect real-world spelling and sentence structure.
- Improved output: Firefly Image 3 also incorporates significant improvements to illustrations in outputs to enable the quick creation of icons, logos, raster images, and line art.
These enhancements to Firefly’s foundation model reflect the competitive nature of the image-generation market, in terms of being able to quickly generate images that are sharp, detail-rich, and, when requested, reflective of the real world in which we live. However, from a commercial perspective, these enhancements are designed to help creators and other workers involved in the content supply chain drive results and value more quickly, with less effort.
As generative AI becomes more commonplace and less expensive to use (due to tech advances and economies of scale), organizations will demand that any creative, generative AI solution is free of unnecessary friction, integrated into workflows that may cross departmental or even company boundaries, and is powerful enough to turn ideas into reality without requiring significant iterations. That is where more powerful foundation models can deliver significant benefit – turning prompts into near production-ready assets quickly.
Transforming Ideation to Creation with New Levels of Creative Control
In addition to the announcement around the Firefly Image 3 Model, Adobe announced a number of innovations and enhancements to its flagship photo-edition application, Photoshop. Now in beta, Photoshop incorporates the following new generative AI-powered features designed to provide creators with more control and more streamlined workflows:
- Reference Image helps to speed image generation by leveraging user-selected images as the source data for new generations.
- Text to Image with Generate Image offers full text-to-image capabilities directly within Photoshop.
- Generate Background replaces and creates backgrounds using prompt language, making it easier than ever to generate content that seamlessly blends into existing images.
- Generate Similar lets creators iterate with variations they select with more precise control.
- Enhance Detail fine-tunes images to improve sharpness and clarity.
Other enhancements to Photoshop include Adjustment Brush, which allows users to apply non-destructive adjustments to specific portions of images; an improved font browser embedded within the application allowing access to more than 25,000 fonts; and Adjustment Presets, which allows users to change images with filters that apply effects in a single click, and with the ability to create and save customized presets.
Improving Efficiency of Creators through GenAI and Embedded Tools
As the demand for content explodes due to the increasing use of social, web, SMS, and other digital channels, creators are being tasked with creating more original content that can be used across each of these media and personalized to a very granular level. As much as the enhancements in Photoshop allow the creation of even more detailed and impressive imagery, the key benefits are efficiency, ease, and speed.
Incorporating these tools into Photoshop addresses these issues while satisfying creators’ desire to generate high-quality images and content that span the gamut from the realistic to the fantastical, along with every degree in between.
It is no secret that other content-creation tools are also implementing new text-to-image generation tools that rival or potentially exceed the capabilities offered by Adobe. However, Adobe’s core strength still lies in its ability to provide high-quality image outputs within the framework of enabling enterprise-grade workflows and IP safety. While these platforms – like any software – may take turns leapfrogging each other in terms of technical capabilities, commercial buyers ultimately should and likely will be focused on the solutions that support efficiency and content authenticity.
Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.
Other Insights from The Futurum Group:
Incorporating Generated Images into Adobe’s Firefly Model
Adobe Posts Solid Q1 FY 2024 Quarter on $5.18B in Revenue
Adobe Delivers Digital Experiences via E-Commerce and Workflow Apps
Author Information
Keith has over 25 years of experience in research, marketing, and consulting-based fields.
He has authored in-depth reports and market forecast studies covering artificial intelligence, biometrics, data analytics, robotics, high performance computing, and quantum computing, with a specific focus on the use of these technologies within large enterprise organizations and SMBs. He has also established strong working relationships with the international technology vendor community and is a frequent speaker at industry conferences and events.
In his career as a financial and technology journalist he has written for national and trade publications, including BusinessWeek, CNBC.com, Investment Dealers’ Digest, The Red Herring, The Communications of the ACM, and Mobile Computing & Communications, among others.
He is a member of the Association of Independent Information Professionals (AIIP).
Keith holds dual Bachelor of Arts degrees in Magazine Journalism and Sociology from Syracuse University.
