Adobe last April Creative cloud introduced the Firefly video model, which will provide video and audio editing with support for artificial intelligence in its applications. Now he has published a video introducing the new generation of artificial intelligence video tools built on Firefly, which can produce video clips from still images.

The new trailer includes features that Adobe announced earlier this year, but were not shown at the time. Of fireflies It featured images produced with the ability to create video from text. The tool allows users to create video clips using text descriptions and customize the results with a variety of “camera controls” that simulate camera angles, movement, and shooting distance. The promotional video also noted that the tool can also create videos from reference images.

Here is the promotional video

Looking at the video, it is noteworthy that the quality of the videos produced parallels OpenAI’s Sora model. It is known that Adobe has previously explored this model for third-party integration of Premiere Pro. On the other hand, Adobe’s tool is currentlyonly 5 seconds can make pictures.

One point where Firefly can gain an advantage over Sora is also specifically emphasized in the promotion. Adobe restricts the use of Firefly is commercially “safe” says. According to the statement, the company used openly licensed, public domain, or licensed images from Adobe Stock to train this model. This eliminates copyright issues.

Adobe Announces New Firefly Video Model Creative cloud, Experience the cloud And Adobe Express states that it will be integrated into its applications. It can also expand the viewable area of ​​existing videos, similar to the Generative Expand feature used for backgrounds in Photoshop. Generative extension The future of the position is also highlighted.

Adobe uses these features towards the end of the year He also indicates that he will make these available.