Uthana Video-to-Motion v2 available on Scenario

Scenario integrates Uthana Video-to-Motion v2 to generate 3D animations from reference videos, allowing creators to export FBX or GLB files without mocap.

As of late April 2026, the Scenario application directly integrates Uthana Video-to-Motion v2, a model that generates 3D animation from a simple reference video, without resorting to a motion capture solution. Scenario, a platform specializing in the creation of 3D assets, images, videos, and audio for studios and independent creators, hosts the tool from Uthana, a company dedicated to generating human motion for video games, animation, and cinema.

The workflow consists of two inputs: a reference video (filmed with a phone, extracted from YouTube, or screen-captured, lasting 2 to 60 seconds, ideally featuring a single person, with a fixed camera, and the full body visible) and a 3D character created or imported into Scenario. The model analyzes the sequence, extracts the 3D motion, and automatically retargets it onto the character's skeleton via an IK retargeting system capable of handling different body types. The animation can be previewed in the interface, then exported as FBX or GLB with full keyframes, ready for Unity, Unreal, Blender, Maya, or Roblox.

A Text-to-Motion variant operates on the same principle, starting from a written description, useful when filming a movement is more complex than describing it. The model is also accessible as a standalone application and via API on Uthana's website, with a batch function to process multiple characters simultaneously. The proposition for independent creators and small studios: substitute a few minutes of iteration for a mocap studio and several weeks of keyframing.