Creating a video based on text or an image? The Adobe Firefly Video Model makes this possible for many users.
Within the Adobe software, Firefly Image allows users to generate an image based on a text prompt. Firefly has been supporting creatives for about a year and a half, including through Firefly Vector (for vector images) and Firefly Design (for template design). Now, its capabilities are being significantly expanded with the launch of the “first publicly available video model designed to be safe for commercial use.”
The video template is available as a beta in the Firefly web app. The model can create a short video clip within minutes through a text prompt, such as in the image below, where the prompt was: “Miniature adorable monsters made out of wool and felt, dancing with each other, 3D render, octane, soft lighting, dreamy bokeh, cinematic.”
Capacities
The Adobe tool offers several options for editing the images, including adjustments for viewing angle, motion, and zooming. Reference images for B-roll generation can also be used to fill gaps in a video timeline. If desired, users can also employ the image-to-video option, which turns a still image into a clip.
Adobe intends to gather feedback from creative professionals in the beta, to which a limited group has access. The software vendor will use this feedback to refine the model further.
Adobe used a dataset of licensed content and publicly available material for previous Firefly features. This approach appears to be continuing with the new video model. The clips now shown are short, around 5 to 10 seconds.
Tip: Adobe adds generative AI to its Creative Cloud platform