DyCraft, a system for animated 3D scene generation that breaks animation into steps that coding agents can more reliably solve, substantially reduces physical violations and generates higher-quality animated 3D scenes that are more often preferred by human raters.
Abstract
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.
Video generation models synthesize visual sequences from text and image inputs. Extending video generation to long animations requires actions, dialogue, camera coverage, and narrative developments to remain coordinated across scenes. Executable scene programs make these decisions available for inspection. Constructing...
Text2Sim is presented, a simulation-specialized agentic pipeline that converts a text-only request into an executable, editable dynamic case and achieves higher scores than all four state-of-the-art baselines on both metrics.
Xiao-Yu Xiong, T. Wang, Yi-Ling Qiao et al.· arXiv.org· 1 citation
Experiments show that Fysiverse-3D-SimReady improves input-view alignment, contact plausibility, and physical stability over existing single-image reconstruction and scene generation baselines, while enabling goal-conditioned physical interactions from a single image.
Lin-Tao Wang, Ming-Yang Sun, Yang Liu et al.· arXiv.org· 1 citation
Controllable world simulation requires more than visually plausible frames: scenes must retain persistent state, coordinated actions, and editable temporal structure. Generative world models offer flexible visual synthesis, but their outputs do not always expose the objects and transitions needed for precise revision....
Zi-Fan Song, Caihua Shan, Kai-Tao Song et al.· 0 citations
Building interactive simulators and scalable robot-learning environments requires a large number of articulated assets. However, most existing 3D assets used in simulation are rigid, and manually converting them into articulated objects with executable kinematic structures is highly labor-intensive, time-consuming, and...
Yi-Xuan Yang, Lu-Yang Xie, Zhen Luo et al.· Proceedings of the 34th ACM...· 4 citations
SceneAgent is presented, an agentic pipeline that coordinates these steps through simulator-based visual inspection and repair and combines object segmentation, physical parameter estimation, part decomposition, articulation, and deformable simulation and demonstrates diverse simulated interactions and preliminary poli...
Luke Hollis, Tian-Xing Fan, Heng Yang· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.