On October 14, 2024, Adobe presented a series of updates to its Firefly AI platform at the annual Adobe MAX conference. Firefly, which has already generated more than 13 billion images since March 2023, is now being expanded to include a commercially secure video model designed to support creative workflows in video editing. This expansion, along with improvements in the areas of image, vector and design models, demonstrates Adobe’s commitment to further developing generative AI as an integral part of creative processes.
Firefly for video: new possibilities for AI-supported video editing
One of the biggest announcements was the introduction of the Firefly Video Model, which is currently in a public beta phase. This new feature makes it possible to generate video clips from simple text or image input. Functions such as “Text to Video” and “Image to Video” offer the possibility to create visual content from text descriptions and to transform still images into dynamic videos. The new Generative Extend feature automatically fills in gaps in the footage, thus facilitating the editing process.
Enhanced image generation and vector tools
In addition to the innovations in the video area, the image processing functions of Firefly have also been improved. With the new Firefly Image 3 model, images are now generated four times faster than before, which is particularly beneficial for users who need high-quality content in a short time. In Photoshop, the new Generative Workspace feature enables faster development of visual concepts.
New features have also been integrated into Illustrator: vector tools such as Generative Shape Fill, Generative Recolor and Text to Pattern give designers more control over the design of vector elements and patterns, allowing for more precise customization to design requirements.
Firefly in the enterprise: AI solutions for professional workflows
Another highlight of the announcements concerns the use of Firefly in companies. New features such as automatic dubbing and lip sync for multilingual video content, as well as the bulk create tool for efficiently editing large volumes of images, are designed to support teams that create content on a large scale. Companies such as PepsiCo, IBM and Deloitte are already using Firefly to improve their creative workflows.
Responsible AI development at Adobe
According to Adobe, the Firefly models are based on licensed and public domain content and developed according to strict ethical guidelines. Particular emphasis is placed on the transparency of the generated content, which is guaranteed by the Content Credentials Initiative – an approach that makes it possible to trace the origin and processing steps of digital works.
Firefly as a platform for creative teams
With the new updates, Adobe Firefly is expanding the possibilities for the use of generative AI and enabling users to create content more efficiently. In the areas of image editing, video production and vector graphics, Firefly offers tools that facilitate creative processes. In particular, the introduction of the video template could change the way videos are produced and edited.
Even though the beta version of Firefly Video is initially only available to a limited group of users, it shows that Adobe is taking a step towards an AI-integrated creative landscape. The future of creative work could be more characterized by collaboration between humans and machines.
Source:
https://news.adobe.com/news/2024/10/101424-adobe-launches-firefly-video-model
