DaVinci MagiHuman

DaVinci MagiHuman

DaVinci MagiHuman makes human-centered video from text descriptions or a reference photo paired with text. Its focus is human performance, including movement, expressions and interactions. In image-based animation, the photo supplies the person's appearance and the scene's visual arrangement, while the description establishes the action.

Both paths accept an optional audio track to guide rhythm, mood and pacing. That gives a soundtrack a role in shaping the generated scene alongside the written direction and, when supplied, the reference image.

What it's good for

The model suits character-driven pitch material and animated portraits. A written narrative moment can become a scene preview, or an established character image can become the starting point for a moving performance.

Marketing teams can develop promotional footage featuring people. Music-video creators can explore portrait sequences shaped around a track, giving editors material for discussing how character movement and musical pacing fit together.