Metraipolis › archive › dc_a7e2124d5865
DyCraft: An Agentic System for Long-Horizon 4D Scene Generation — arXiv:2610.40754
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.40754
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T05:45:40.639Z
- SHA-256 of the archived text
7e5766ea1269d829307e30030d702a639c335dee61b9c4fab86d2cf2c1f99a6f- Stable URL of this page
https://metraipolis.com/documento/dc_a7e2124d5865
How to cite
DyCraft: An Agentic System for Long-Horizon 4D Scene Generation — arXiv:2610.40754 — arXiv (en), https://arxiv.org/abs/2610.40754 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a7e2124d5865 (Metraipolis archive dc_a7e2124d5865, sha256 7e5766ea1269d829307e30030d702a639c335dee61b9c4fab86d2cf2c1f99a6f)
BibTeX
@misc{dca7e2124d5865,
title = {DyCraft: An Agentic System for Long-Horizon 4D Scene Generation — arXiv:2610.40754},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.40754},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a7e2124d5865; sha256 7e5766ea1269d829307e30030d702a639c335dee61b9c4fab86d2cf2c1f99a6f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40754. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2610.40754 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T05:33:32.257Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).