跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 25 天前AI 評分44

DyCraft:用智能體生成長時程 4D 場景的系統

DyCraft: An Agentic System for Long-Horizon 4D Scene Generation

AI 導讀

DyCraft 將動畫製作拆分為靜態場景、關鍵幀故事板和局部動作編排,以減少長時程 3D 動畫中的物理錯誤。系統在 24 項長時程、多鏡頭動畫任務中接受評估;與 Codex 中原生 GPT-6 智能體及先前的智能體基線相比,它減少了物理違規,生成的動畫 3D 場景品質較高,也更常獲人類評審選擇。

正文

Published on Oct 29

Authors:

,

,

,

,

,

,

Abstract

DyCraft improves long-horizon 3D animation by decomposing it into planned static scenes, keyframe storyboards, and local motion authoring to reduce physical errors.

While frontier coding agents can generate high-quality 3D scenes and articulated assets, they struggle to create coherent long-horizon animations. When equipped with general-purpose harnesses, they frequently produce physically implausible outputs, generating objects that interpenetrate, lack physical support, or move unnaturally. Such bare agents struggle to repair these errors because they default to scripting the full animation in one pass, making it difficult to verify their work and disentangle complex spatiotemporal constraints. Drawing on the practice of human animators, we introduce DyCraft, a system for animated 3D scene generation that breaks animation into steps that coding agents can more reliably solve. DyCraft first expands the user's prompt into a detailed plan and builds a static scene. It then creates a 3D storyboard of key animation poses, where each frame is a static 3D scene that the agent can build and debug in isolation. Storyboarding decomposes the problem into pairs of verified start and end states, allowing the agent to author local motion between them with a collection of motion authoring tools. We evaluate DyCraft on 24 long-horizon, multi-shot animation tasks covering humanoid robots, human and animal characters, and physical mechanisms. Compared with a native GPT-6 agent in Codex and prior agentic baselines, DyCraft substantially reduces physical violations and generates higher-quality animated 3D scenes that are more often preferred by human raters.

View arXiv page View PDF Project page GitHub 11 Add to collection

Get this paper in your agent:

hf papers read 2610.40754

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.40754 in a model README.md to link it from this page.

Datasets citing this paper 2

jefftan969/dycraft-films

Updated 2 days ago

•

103

jefftan969/dycraft-assets

Updated 2 days ago

•

42

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.40754 in a Space README.md to link it from this page.

Collections including this paper 1

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co