Apple has expanded its range of artificial intelligence tools by introducing a new tool that utilizes large language models (LLMs) to animate static images based on user-provided text prompts.
The latest development from Apple is showcased in a research paper titled “Keyframer: Empowering Animation Design Using Large Language Models,” where they introduce a tool that leverages the capabilities of LLMs to generate animations from static images through natural language input.
Unlike traditional text-to-image systems like DALL·E and Midjourney, which are commonly used for static images, animations necessitate a more intricate set of user inputs such as timing and coordination. Apple’s Keyframer addresses this complexity by allowing users to iteratively refine their animated designs through a sequence of prompts, enabling them to have more creative control over the process.
By combining design principles for language-based input with the code-generation abilities of LLMs, Keyframer generates CSS animation code to bring static 2D images, such as Scalable Vector Graphics (SVG), to life. Users can upload an SVG image and input prompts like “generate three designs where the sky fades into different colors and the stars twinkle,” prompting Keyframer to produce the corresponding CSS animation code.
The research behind Keyframer was influenced by insights from professional animation designers and engineers, aiming to streamline the design prototyping process and empower creators to experiment with dynamic editing features while harnessing the generative capabilities of LLMs.
Apple’s continuous advancements in AI technology also include recent breakthroughs such as an AI model for pixel-level image edits using multimodal LLMs, as well as innovative techniques to deploy LLMs on memory-constrained Apple devices.
Reports from The Information and analyst Jeff Pu suggest that Apple is poised to introduce generative AI features on the iPhone and iPad with the upcoming iOS 18 update. This update is anticipated to enhance Siri with ChatGPT-like generative AI functionality, marking a significant milestone in the history of Apple’s mobile software evolution.
This latest innovation from Apple underscores their commitment to pushing the boundaries of AI technology and providing users with cutting-edge tools for creative expression and design exploration.
(Source: VentureBeat)
