Sept 11 (Askume) – Adobe (ADBE.O) said on Wednesday the software maker will launch a new artificial intelligence video creation and editing tool later this year focused on the needs of creative professionals.
The artificial intelligence tool, called the AdAdobe Firefly Video Model, will be released in beta and joins the Photoshop maker’s existing range of Firefly image-generating apps, allowing users to create static images, designs and vector graphics.
This model would give Adobe access to the growing market of AI-based video creation tools, which OpenAI’s Sora, Stability AI’s Stable Video Diffusion, and other smaller startups are already targeting with AI video apps.
The tool can produce five-second clips of a single cue and annotate text and image cues, said Alexandru Costin, vice president of generative artificial intelligence at Adobe. Users can also specify the desired camera angle, pan, movement and zoom.
“We’ve invested in bringing this model to the level of quality and instant understanding that photographers expect, and we’ve invested in making sure that we really pay attention to the signal … and unlike other (artificial intelligence video) models, respect the photographer’s guidance over comparison,” Costin told Askume in an interview.
Adobe says the video model was trained on public domain or licensed content it has the rights to use, and not on any Adobe customer content.
“We train them entirely on content from the Adobe Stock library, which includes 400 million images, illustrations and videos and contains no intellectual property, trademarks or identifiable characters,” Costin said.
Adobe is also launching a tool called Generative Extend, which will be available in its Premiere Pro video editing software, that can extend any existing clip by up to two seconds by creating suitable inserts to fill gaps in the footage.
The tool was first previewed in April and has received “overwhelmingly positive feedback from all of our customers,” Kostin said.
