At its MAX 2024 conference, Adobe announced the Firefly video model to compete with OpenAI and Meta. The company said its new generative AI model allows creators to generate videos from text and image prompts. Additionally, Adobe announced improvements to its other AI models as well as new features for Photoshop, Premiere Pro, Illustrator, and more. Here are the details:
Adobe Firefly video template in Firefly web app and Premiere Pro
Click here to join us on WhatsApp
With its Firefly video template, Adobe is expanding its family of generative AI models, which already includes an image model, a vector model, and a design model. The Firefly video template is now available on the Firefly web app as text-to-video and image-to-video features in limited public beta. The text-to-video feature allows users to generate a video from text prompts. It also provides access to various camera controls, such as angle, movement and zoom, to fine-tune videos. Picture-to-video capabilities allow creators to bring still images or illustrations to life by turning them into video clips.
Adobe said the Firefly video template is deeply integrated into its products, such as Premiere Pro, offering more advanced features such as Generative Extend. This new feature allows creators to extend clips to fill in gaps in footage and smooth transitions.
Photoshop
Powered by Firefly’s Image 3 template, which Adobe says generates images four times faster, Photoshop introduces a new generative workspace feature. This feature allows creators to imagine, brainstorm, and rehearse concepts simultaneously. Other new features include Distraction Removal, which removes unwanted objects from images, and a new Substance 3D viewer that allows viewing and editing 3D objects while working with 2D designs.
Illustrator
Adobe Illustrator is also receiving new AI-based features from the existing Firefly vector model. This includes a new Objects on Path feature that allows users to attach, arrange, and move an object along any path on their workbench. Additionally, there is a new Enhanced Image Trace feature that makes converting images to vectors simpler and faster.
First publication: October 15, 2024 | 12:40 p.m. STI
