Adobe’s working on generative video

17.04.2024

 

Adobe is venturing into the realm of AI-driven video generation, joining the ranks of OpenAI’s Sora and Google’s Imagen 2. This forthcoming addition to Adobe’s Firefly suite, scheduled for integration into Premiere Pro later in the year, represents the company’s strategic move to compete in the rapidly evolving landscape of generative AI video tools.

The core functionalities of Adobe’s AI model will encompass object addition, removal, and generative extend, each tailored to enhance the video editing process with unprecedented ease and creativity. Object addition empowers users to seamlessly insert objects into specific segments of video clips, leveraging either prompts or reference images to guide the AI’s rendering process. As an illustrative example, Adobe showcased the generation of a briefcase brimming with diamonds, demonstrating the model’s ability to seamlessly integrate virtual elements into real-world footage.

Conversely, object removal functionality equips editors with the means to effortlessly eliminate unwanted elements from scenes, whether it be a pesky boom mic intruding into the frame or a stray coffee cup disrupting the visual composition. This feature underscores Adobe’s commitment to streamlining the editing workflow by automating tedious tasks that would traditionally demand meticulous manual intervention.

Furthermore, the generative extend feature represents a novel approach to extending the duration of video clips with a touch of artistic finesse. By adding frames to the beginning or end of clips, editors gain greater flexibility in fine-tuning pacing and synchronization with accompanying audio tracks. While Adobe has yet to disclose the precise number of frames added through this process, the primary aim is to provide editors with a nuanced tool for orchestrating seamless transitions and imbuing shots with added emotional depth.

In essence, Adobe’s foray into AI-driven video generation signifies a paradigm shift in the realm of digital content creation, where cutting-edge technologies converge to empower creators with unprecedented capabilities. As the boundaries of what is achievable continue to expand, Adobe remains at the forefront of innovation, continually pushing the envelope to redefine the possibilities of video editing and storytelling in the digital age.

Other news

Meta to add ‘AI generated’ label to images created with OpenAI, Midjourney and other tools

Meta says it’s working to identify and label AI-generated images shared on its platforms that were created by third-party tools, as the company prepares for the 2024 election season amid a proliferation of artificial intelligence tools that threaten to muddy the information ecosystem.

Read More

Landmark Collaboration: UK-Canada Pact on AI Computing Unveiled

The UK and Canada have recently inked a groundbreaking agreement aimed at fostering collaboration in the realm of computing power to propel advancements in AI research and development.

Read More
en_USEnglish