Runway launches Act-One for character animation
Runway has released Act-One, a Gen-3 Alpha tool that turns one actor's video and voice performance into animated or live-action characters without motion-capture equipment or manual face rigging.
Runway announced Act-One, a new tool inside its Gen-3 Alpha video model that generates character performances from video and voice inputs. The company calls it "state-of-the-art," but the announcement includes no benchmarks or technical details about the model.
Traditional facial animation pipelines often rely on motion-capture equipment, multiple footage references and manual face rigging to move an actor's performance onto a 3D model. Runway says Act-One skips those steps. It works "directly and only" from footage of an actor's performance and needs no extra equipment.
According to Runway, the model keeps realistic facial expressions when it maps a performance onto characters whose proportions differ from the source video. The company also says it holds up across camera angles in cinematic, live-action output. Runway highlights multi-turn dialogue scenes, which it says were previously hard to produce with generative video models. In its example, a creator uses a consumer-grade camera and a single actor who performs several characters from a script.
On safety, Runway lists three measures: stronger detection and blocking of attempts to generate content showing public figures, technical checks that a user has the right to use any voice they create with its Custom Voices feature, and continuous monitoring for other misuse.
Act-One is now available in Gen-3 Alpha. The announcement doesn't cover pricing, plan tiers, output length limits or how the model was trained.