# Runway Act-Two Canonical page: https://fuser.studio/models/runway-act-two Runway Act-Two in Fuser drives a character image or video with the face and body performance from a reference video. Build talking scenes on one canvas. ## Specs | Spec | Value | | --- | --- | | Inputs | Character Video (video), Character Image (image), Reference Video (video) | | Output | video | | Aspect ratios | 1280:720 Horizontal, 720:1280 Vertical, 960:960 Square, 1104:832 Horizontal, 832:1104 Vertical, 1584:672 Landscape | | Cost | 69–4,133 credits per run | Transfer expressive human performance, dialogue lip-sync, and nuanced upper-body gestures directly onto custom character art and video Creator: Runway [Animate performance](https://app.fuser.studio) ## How Runway Act-Two works Upload your character asset, record your driving performance on any standard camera, and calibrate expression intensity to generate fluid character video. ### Provide your character asset Upload a front or three-quarter image or video of your character, keeping hands and upper body in frame for full performance tracking. ### Record driving performance footage Supply a clear 3 to 30 second webcam or camera clip of an actor delivering speech, head turns, and hand gestures. ### Tune expression intensity and render Adjust the expression intensity slider from subtle realism to theatrical animation, toggle body control, and generate your video. ## What Runway Act-Two is good at Explore full-body gesture tracking, intensity scaling, and dynamic environmental lighting designed for rapid narrative animation. ### Full-Body Gesture Transfer Map expressive hand movement, torso posture, and head turns directly from driving reference footage onto stylized 2D illustrations or 3D renders with automatic environmental lighting. ### Fine-Grained Expression Scaling Control emotional delivery on a calibrated 1 to 5 intensity scale, tailoring micro-expressions from understated photorealism to bold, exaggerated animation. ### Dynamic Secondary Motion Automatically generate cohesive secondary movement across hair, clothing, and cast shadows that dynamically adjust to the character's rotational momentum. ### Flexible Duration Window Animate sequences ranging from 3 to 30 seconds using single-camera driving clips without multi-rig motion capture suits or manual facial rigging. ## Made with Runway Act-Two Discover how creators map human performance across 3D renders, physical sculptures, 2D sketches, and stylized digital avatars. ### Precision performance mapping on polished metallic characters ### Textured puppet character with expressive speech delivery ### Articulated wooden figure animated from webcam reference ### Dynamic 2D charcoal sketch brought to life with full dialogue capture ### Stylized low-poly avatar driven by mobile camera video ## What people build with Runway Act-Two See how directors, game developers, fashion studios, and animators use performance transfer to accelerate character production. ### Cinematic Narrative Pre-Visualization Block complex character dialogues and emotional beats in pre-production without staging multi-actor motion capture sessions. ### Indie Game Character Prototyping Test dialogue animations, NPC idle gestures, and narrative cutscenes quickly by recording driving performances straight from a webcam. ### High-Fashion Lookbook Motion Bring static editorial portraits and avant-garde character concepts to life with fluid posture, subtle head tilts, and wardrobe physics. ### Brand Commercials and Virtual Avatars Produce talking-head brand ambassadors, digital mascots, and social video hosts matching human vocal cadence and hand gestures. ### Animated Storyboarding and Comic Dubs Transform static 2D illustrations and graphic novel panels into moving dialogue sequences with synchronized speech and expressive acting. ## Frequently Asked Questions ### What inputs do I need to animate a character with Act-Two? You need a driving reference video and either a character image or character video. The reference video should be a 3 to 30-second clip of an actor with clearly visible facial features and upper body under even lighting. Your character reference can be a 2D illustration, 3D render, clay model, or photorealistic image. ### How does the Expression Intensity parameter work? Expression Intensity is a control from 1 to 5 that dictates how dramatically facial movements transfer. Lower settings (1 or 2) are recommended for photorealistic humans to keep lip-sync natural and prevent skin distortion. Higher settings (4 or 5) suit cartoon characters, monsters, or stylized avatars where exaggerated expressions look natural. ### When should I enable or disable Body Control? Enable Body Control when your character asset and driving video both clearly show the upper torso and hands at rest. This transfers hand gestures, posture shifts, and head movement. If your character asset is a tight headshot without visible hands, keep Body Control disabled to avoid distorted hand generation. ### What types of motion should I avoid with this model? Avoid fast action choreography, martial arts, combat sequences, and scenes requiring characters to hold or interact with handheld props like phones or weapons. Act-Two is designed for dialogue and upper-body gestures; rapid physical movements or full sideways profile angles cause tracking loss. ### What are the duration limits and resolution options? Driving reference videos must run between 3 and 30 seconds in duration. Act-Two supports six native aspect ratios: 1280:720 horizontal, 720:1280 vertical, 960:960 square, 1104:832 horizontal, 832:1104 vertical, and 1584:672 widescreen landscape. ## Try Runway Act-Two on Fuser Transfer expressive human performance, dialogue lip-sync, and nuanced upper-body gestures directly onto custom character art and video [Animate performance](https://app.fuser.studio) ## Related models - [Runway Aleph](https://fuser.studio/models/runway-aleph.md) - [Runway Video](https://fuser.studio/models/runway-video.md) - [MiniMax H3 Recast](https://fuser.studio/models/minimax-h3-recast.md) - [Kling O3 Edit](https://fuser.studio/models/kling-o3-edit.md) - [Kling O3](https://fuser.studio/models/kling-o3.md) - [Kling 3.0 Video](https://fuser.studio/models/kling-3-0-video.md) Node reference: [Runway Act-Two docs](https://docs.fuser.studio/docs/nodes/video/runway-act-two.md) ## More articles - [How to Replace People in a Video With AI: Recast, Kling O3 Edit and Motion Transfer](https://fuser.studio/articles/replace-people-in-video-ai.md) - [Best AI Lip Sync Tools (2026): One Voice Line, Four Models](https://fuser.studio/articles/best-ai-lip-sync-tools.md) - [Best AI Avatar Video Generators: One Portrait, One Script, Four Models](https://fuser.studio/articles/best-ai-avatar-video-generators.md)