New Apple Keyframer AI tool can create animated clip from still images

Apple recently published a new large language model (LLM) application named Keyframer that can animate still images based on user prompts using artificial intelligence (AI).

Information about this Keyframer AI tool has been published in the research paper titled “Empowering Animation Design using Large Language Models”. It aims to explore more ways to describe motion in natural language.

The feature works quite futuristic and you can create an entire scenario based on one image.

For example, if you are an animator with a single image drawn of a character and a story script. The Keyframer would use the script to generate a story featuring that single character image, all animated.

This could be an amazing capability and evolution among generative AI applications. You can check the image below and how it was composed out of some references.

Apple Still Image Animator AI Tool – Keyframer

So far, we’ve seen generative AI’s potential in natural language processing but adding it to animations would take the LLM to the next level.

Keyframer uses LLM which generates CSS animation code from a static SVG image and user prompt. As an AI technology, there might be some improvements required in editing and quality aspects to match the user’s expectations.

(source – VentureBeat)