Adobe is set to unveil a new generative AI-powered video creation and editing tool later this year in a limited beta release. Named Adobe Firefly Video Model, the tool will expand the company’s suite of AI-driven applications, joining the Firefly family that already supports image, design, and vector graphic generation.
This new video tool marks Adobe’s entry into the competitive field of AI-based video generation, a space currently populated by platforms like OpenAI’s Sora and Stability AI’s Stable Video Diffusion, Reuters news report said.
Adobe said its Firefly Video Model can generate five-second video clips based on text or image prompts, allowing users to specify elements such as camera angles, motion, and zoom. The tool is aimed at meeting the high standards of professional videographers, according to Alexandru Costin, Adobe’s vice president of generative AI.
Alexandru Costin said the AI tool will have the ability to better understand user prompts and respect videographers’ input. He also confirmed that the model has been trained exclusively on public domain or licensed content, avoiding intellectual property issues by using the Adobe Stock database, which contains over 400 million curated images, illustrations, and videos.
Along with Firefly Video Model, Adobe is introducing Generative Extend, a feature in Premiere Pro that can automatically add two seconds of generated content to extend any video clip, filling in gaps in footage seamlessly. Both tools aim to enhance Adobe’s offerings for creative professionals, as the company continues to leverage AI to streamline the video editing and production process.
Adobe Firefly Video Model better for video production than ChatGPT?
InfotechLead.com Research indicates Adobe’s Firefly Video Model is aiming to revolutionize the video production industry, far surpassing what generative AI models like ChatGPT can offer for visual storytelling. While ChatGPT excels in generating text-based content and ideas, Adobe’s Firefly tools are tailored specifically for filmmakers, video editors, and creative professionals, providing them with unprecedented levels of control and creativity in video editing.
Adobe previewed advancements to its Firefly Video Model, which extends the company’s suite of generative AI models, already responsible for producing over 12 billion images globally. Firefly’s latest update introduces capabilities that go beyond text, empowering users to generate entire video clips from simple text prompts, transform images into live-action footage, and seamlessly integrate effects and animations within existing projects.
ChatGPT is optimized for natural language generation and conversational AI. Adobe Firefly Video focuses on enhancing video creation workflows. It allows users to finetune elements like camera angle, motion, and zoom within a video, providing a level of visual precision unmatched by text-based AI systems.
The integration of Firefly’s Text to Video and Image to Video features directly into Adobe Premiere Pro, a leading video editing software, allows editors to fill gaps in timelines and add new elements without the need for extensive reshoots. By referencing text or images, creators can automatically generate B-roll or motion graphics, tasks that typically require manual editing or advanced visual effects software. This automation significantly reduces production time, giving professionals more room to focus on storytelling and creativity.
ChatGPT is a valuable tool for brainstorming ideas, scripting, and aiding in content creation for written or spoken media. However, Adobe’s generative AI tools, designed specifically for visual content, provide a more robust solution for video professionals by offering a greater range of customization options, seamless integration into professional editing workflows, and industry-grade commercial safety through its trained models.
Baburajan Kizhakedath
