Unofficial AI-summarized news site (not affiliated with any AI company)
AI News JP / www.ai-news.jp
🟠 Important AI Summary · Source: Runway

Animation production is being revolutionized with the new tool Act-One!

Runway Introduces Act-One: A New Tool for Character Animation

Original: Runway Research | Introducing Act-One

Importance: 新たなアニメーションツールの発表は多くのクリエイターに影響を与えるため。

Summary

Runway has announced a new tool called Act-One, which generates expressive character performances using video and voice inputs. It simplifies traditional animation workflows by directly converting an actor's performance into a 3D model, preserving realistic facial expressions while allowing for diverse character designs. Notably, Act-One enables narrative content creation using just a consumer-grade camera and one actor. The tool also includes content moderation features to ensure safety.

Key Points

  • Simplified animation production workflow
  • Transforms actual performances into 3D models
  • Enables diverse character design
  • Generates narrative content with a consumer camera
  • Includes safety measures and content moderation features
View developer notes (APIs, breaking changes, migration)

Act-One is a new character animation tool for Gen-3 Alpha that uses video and voice inputs. It eliminates the need for traditional motion capture and manual face rigging, directly converting an actor's performance into a 3D model. It can accommodate various reference images while preserving realistic facial expressions and producing characters with different proportions from the source video. The tool also maintains high robustness across camera angles for realistic output. This enables the generation of multi-turn dialogue scenes, previously challenging with generative video models, and includes a comprehensive suite of content moderation and safety measures.

モデル新機能安全性/研究Audience: 一般ユーザーAudience: 開発者

Source: https://runway.com/research/introducing-act-one

Outlet: Runway

This article is an AI-generated summary (OpenAI GPT-4o-mini) of publicly available information from Anthropic, OpenAI, Google, Meta, Mistral, DeepSeek, Sakana, and other vendors. The original source URL is always provided in accordance with fair-use citation requirements. Summaries are AI-generated and may contain mistranslations or misinterpretations. Always verify details with the original source.