Apple is making another addition to its AI basket with a new tool that enables users to create animation from static images as the AI race intensifies.
The tool, Apple Keyframer, leverages large language models (LLM) and only requires a text prompt from users for it to generate animation. This project was reportedly informed by insights from interviews done with engineers and professional animation designers.
Apple claims its tool is more powerful
A research paper titled “Keyframer: Empowering Animation Using Large Language Models” details the latest AI tool, whose initial target market is graphics professionals.
According to the researchers, unlike other generative AI tools like Dall-E and Midjourney, animations require “a more complex set of user considerations.” These, according to the paper, include timing and coordination “that are difficult to fully specify in a single prompt.”
As such, users need alternative approaches that allow them to “iteratively construct” and refine generated designs. This is especially the case when creating animations, according to the research paper.
“We combined emerging design principles for language-based prompting of design artifacts with the code-generation capabilities of LLMs to build a new AI-powered animation tool called Keyframer,” reads part of the paper.
“With Keyframer, users can create animated illustrations from static 2D images via natural language prompting. Using GPT-4 3, Keyframer generates CSS animation code to animate an input Scalable Vector Graphic (SVG).”
Also read: Scientist Integrates AI with Over 500 African Native Languages
Using Apple Keyframer
According to MacRumors, the first stage is uploading an SVG image of anything, for example,e a rocket. The next step is to prompt, “Generate three designs where the sky fades into different colors and the stars twinkle.”
The AI tool will then create a CSS code for the animation. To refine the animation, a user needs to edit the CSS code directly or, alternatively, by typing in more prompts.
According to the MacRumors report, Keyframer provides three options, which users can edit accordingly.
“Keyframer enables users to iteratively refine their designs through sequential prompting, rather than having to consider their entire design upfront,” explain the authors.
“Through this work, we hope to inspire future animation design tools that combine the powerful generative capabilities of LLMs to expedite design prototyping with dynamic editors that enable creators to maintain creative control.”
Wider use
Although the research paper insinuates an initial target market of professional animators and designers, the Apple Keyframer can be used by others too.
Social media influencers and small businesses can take advantage of the tool to create eye-catching video content for their websites or TikTok pages.
But that’s not all. According to 9to5Mac, ordinary users can also derive fun from using the tool, providing another form of filter.
Apple has been criticized for allegedly lagging behind in AI, but the latest development shows the company has been working quietly in its quest to make a mark in the AI industry. The market criticized Siri, saying it had not evolved like other AI assistants from Amazon and Google, although Apple has used AI in other areas for over a decade.
Revelations of Apple Keyframer come a few days after the tech giant also introduced MGI, which is a tool for editing pictures using text prompts.
Although both Keyframer and MGI are not expected to make a major push for the company’s AI agenda as they are still in their early stages, the two are likely to find their way on some Apple devices “further down the line.”
and then