
AI performance capture. Webcam to character animation in seconds.
Record yourself acting a scene on a webcam or phone, upload a character reference image, and Runway Act Two transfers your full performance onto the character. Facial expressions, lip sync, hand gestures, and body posture all carry across. No motion capture suit, no studio, no rigging. The character moves exactly as you did, with your timing and your emotion.

“A stylised cartoon mascot character delivering an enthusiastic product announcement, expressive hand gestures, bright studio lighting, brand content”

“A photorealistic digital human presenting a keynote speech, subtle facial expressions, professional lighting, corporate event setting”

“A medieval knight character delivering dramatic dialogue in a torch-lit castle hall, intense facial performance, cinematic framing”

“An animated character host for a children's educational video, exaggerated friendly expressions, colourful classroom background”

“A concept art robot character mimicking human gestures during a product demonstration, industrial lab setting, sci-fi design”

“A fashion illustration character walking and posing on a runway, editorial movement, stylised proportions, watercolour aesthetic”
Type a detailed prompt describing the video you want, or upload a reference image as a starting frame.
Pick your resolution and duration. See the credit cost before you generate.
Your video is ready in 1-3 minutes. Download, iterate, or extend the sequence.
Jump into the Studio and start generating. Plans from £10/month.
Choose a PlanTraditional character animation requires a motion capture studio, a retargeting pipeline, and hours of cleanup. Runway Act Two replaces all of it with a webcam and a text prompt. Record yourself performing a scene. The model captures your facial micro-expressions, lip and mouth movement, hand gestures, and full-body posture. Upload a character reference image and Act Two generates a video of that character performing your exact movements with your exact timing.
The transfer is not a face swap or a filter. Act Two regenerates the entire scene from scratch, placing the character in a plausible environment with appropriate lighting and physics. If the character is a stylised illustration, the output maintains that style. If it is a photorealistic portrait, the output stays photorealistic. Expression intensity is configurable: dial it up for exaggerated cartoon performances or keep it subtle for realistic drama. Body control settings let you balance between strict motion fidelity and natural character-specific movement.
For design professionals, this opens character animation to anyone with a brief and a phone. Game developers can prototype cutscene performances without rigging a model. Architects can place an animated presenter into a walkthrough video. Brand teams can create character-led social content without hiring voice actors or animators. Film pre-production teams can block out performances for storyboard validation. Each generation takes minutes, not days.
Act Two captures everything: facial expressions down to eyebrow raises and lip curls, hand gestures including individual finger positions, and full-body posture from stance to weight shifts. The transfer preserves your timing, so pauses land where you intended and emphasis hits on the right beat.
The character reference can be a photograph, an illustration, a 3D render, or a concept sketch. Act Two adapts the performance to the character's proportions and visual style. A cartoon mascot moves with cartoon physics. A realistic human moves with human weight. The style of the reference drives the style of the output.
A small indie studio building creative tools the way they should be built. No VC theatre, no funnel games, no faceless support.
Professional video generation. Plans from £10/month.