Adobe Firefly Video Model: Adobe To Launch Its GenAI Video Creation Tool This Year To Rival OpenAI Sora, Stability AI’s Stable Video Diffusion

San Jose, September 12: Adobe, a US-based tech company known for tools such as Adobe PDF Reader, Adobe Photoshop, Adobe Illustrator and Adobe Premier Pro, will introduce its new generative AI-powered video creation tool this year. The latest Adobe genAI video creation tool will allow users to edit the videos. First, the tech company will release a beta version for testing and then introduce it to the suite of its other applications.

According to a report by Reuters, Adobe's new genAI video tool, dubbed 'Adobe Firefly Video Model,' will be released this year as beta and later will be part of Photoshop's maker's current Firefly image-generation applications. With the introduction of this new video generation tool, Adobe will strengthen its suite of applications used by creative professionals. India Will Become World’s Third Largest Economy Due to Its Semiconductor Technology, Says Union Minister Jitin Prasada. 

Adobe Firefly Video Model will take on the existing genAI-powered video generation tools such as OpenAI Sora, Stable Video Diffusion and China's Kling. The report highlighted that the upcoming text-to-video model from Adobe could generate a five-second clip from a single prompt. The prompt given by the users could be in text or image format, which would create a video using generative artificial intelligence. 

Besides video generation, the Firefly Video Model would also allow users to adjust camera angle, zoom, motion, and panning. The company said it would consider the expectations of videographers and develop this model to offer quality and prompt understanding. Alexandru Costin, Vice President of Generative AI and Sensei at Adobe, said that the company invested in making the model deliver high quality and prompt understanding. 

He further said Adobe ensured it paid attention to the prompts and respected the videographers' guidance to make a better AI video model. The model is reportedly trained on the public domain or licensed content with permission. Adobe has not used any content of its customers. The model is trained on 400 million images available on the Adobe Stock data, without intellectual with party issues, trademarks and recon, or recognisable characters. Arya.ai CEO Vinay Kumar Sankarapu Aims to Build Open-Source Community for Safe Super Intelligence Research.

Besides, the company would likely roll out the Generative Expand tool for its Premier Pro editing software. It would extend the clip by two seconds by generating a proper insert and filling the gaps in the video.

(The above story first appeared on LatestLY on Sep 12, 2024 02:11 PM IST. For more news and updates on politics, world, sports, entertainment and lifestyle, log on to our website latestly.com).