In a bold leap forward for AI-driven filmmaking, New York-based startup Runway has launched two powerful new performance capture tools—Act-One and Act-Two—designed to eliminate the need for traditional motion capture equipment and put high-fidelity animation in the hands of anyone with a smartphone camera.

Coming on the heels of its blockbuster partnership with Lionsgate to develop custom generative video models based on the studio’s 20,000-title catalog, Runway’s latest releases mark a new chapter in the rapidly evolving production landscape. Together, Act-One and Act-Two promise to upend not only VFX-heavy filmmaking workflows, but also how studios think about cost, talent, and post-production timelines.

From Face to Frame: Act-One Brings Nuanced Facial MoCap to the Masses

Launched late last year, Act-One quietly set a new standard for AI-assisted facial animation. The tool allows creators to record any actor’s performance—from an iPhone selfie or studio-grade camera—and map the facial expression, gaze, eye-line, and even micro-emotions onto fully generated digital characters. Unlike legacy systems that require elaborate rigs and hours of cleanup, Act-One handles it all via a single video input.

“This is a game changer,” posted one early user on X. Industry insiders agreed.

Act-One’s realism across varied art styles, character designs, and camera angles unlocks a new tier of expressiveness for indie creators, game developers, and studio animation teams alike. In an industry long defined by high-end motion capture rigs, cumbersome facial markers, and Oscar-winning 3D face rigs à la The Curious Case of Benjamin Button, Runway’s tool offers near-cinematic realism without the overhead.

“We’re now beyond the question of whether AI can produce consistent video,” said Runway CEO Cristóbal Valenzuela in a public post. “The baseline is good models. Now the focus is how we apply them.”

Act-Two: Full-Body MoCap, AI-Style

This week, Runway followed up with Act-Two, a next-generation full-body motion capture model. Where Act-One zeroed in on the face, Act-Two brings AI MoCap to the entire body—head, hands, feet, torso—capturing complex movements with unprecedented fidelity.

The company claims Act-Two can animate any character in any environment with a single performance input. Stylized, photoreal, 3D, 2D—no problem. And unlike traditional capture stages, no suits or markers are required. One actor. One performance. Infinite character outputs.

Together, these tools position Runway as the leader in end-to-end AI performance capture—face to feet—with full integration into its existing Gen-3 Alpha platform for text-to-video, video-to-video, and image-to-video generation.

Hollywood Impact: From Backlot to Browser

Runway’s momentum is more than technical—it’s strategic. Their alliance with Lionsgate signals a serious move into studio-grade workflows. By building custom models trained on the studio’s vast IP library, Runway isn’t just offering tools; it’s offering pipelines.

The implications are massive: directors can previsualize or even complete entire performances without booking a stage. Indie filmmakers can cast themselves in every role. VFX houses can bypass costly motion capture sessions. And game studios can generate expressive, full-body performances without animation teams or trackers.

And it’s not just about cost. The technology also scales emotion and nuance. A single actor, with a single take, can now play a dozen characters across styles, worlds, and genres—all while maintaining performance integrity.

Safety First, Innovation Fast

To address growing concerns around digital impersonation, both Act-One and Act-Two come equipped with built-in safeguards. These include the detection and blocking of unauthorized content involving public figures and tools to verify the authenticity of source materials.

Runway continues to emphasize ethical development, echoing its earlier commitment to responsible content generation amid the rapidly growing generative AI sector.

The MoCap Industry Reimagined

In less than a year, Runway has gone from a generative video upstart to a leading force in performance capture. With Act-One and Act-Two now live for enterprise customers and creative partners—and Act-One beginning gradual rollout for free users with credit access—the company is signaling not just a feature drop, but a redefinition of digital storytelling.

As AI models continue to compress cost, time, and complexity from traditional workflows, the legacy of motion capture may no longer rest on ping pong balls and head rigs—but on pixels, algorithms, and creativity.