Tech startup Wonder Dynamics has recently released a groundbreaking AI tool, Wonder Animation Beta, co-founded by Hollywood actor Tye Sheridan. This company is revolutionizing the film production process with artificial intelligence.

This innovative video-to-3D-scene technology allows creators to use any camera and shoot in any location, then convert the footage into a 3D animated scene with CG characters through AI reconstruction. Compared to the company's previous Wonder Studio, the new technology has made a significant leap in functionality.

image.png

The most striking feature of Wonder Animation is its powerful scene reconstruction capability. It can not only handle complex sequences with multiple camera cuts but also accurately restore the positional relationships and movement trajectories between each camera, characters, and the environment in 3D space. This means creators can obtain a complete virtual scene, including all camera settings, character body and facial animations, and all elements (animations, characters, environments, lighting, and camera tracking data) can be freely edited in mainstream software like Maya, Blender, or Unreal Engine.

Nikola Todorovic, co-founder of Wonder Dynamics, said: "Many people mistakenly believe that AI is a one-click solution, but that's not the case. The purpose of releasing the Wonder Animation Beta is to make it easier for artists to create full-animation films while maintaining complete creative control. Unlike the black-box approach of most generative AI tools on the market, we are committed to allowing artists to shape their own vision rather than simply relying on automated outputs."

It's worth noting that Wonder Dynamics was founded in 2017 and was acquired by Autodesk in May this year. This acquisition not only expanded the company's influence but also allowed more creators to achieve Hollywood-level special effects on an independent film budget. The emergence of this technology is expected to break budget constraints, allowing more creative works to be presented.