Metraipolis › archive › dc_cb2b3a2c6524

Movens-Zero: Scaling Humanoid Motion Foundation Models with Large-Scale Human Videos — arXiv:2610.41877

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41877
License (as archived)
CC0-1.0
Archived
2026-11-03T05:47:11.567Z
SHA-256 of the archived text
4fbaa196e62008cbf8701c07ea765402e36edf1fb25d0cf68622d7a68f6b6e80
Stable URL of this page
https://metraipolis.com/documento/dc_cb2b3a2c6524

How to cite

Movens-Zero: Scaling Humanoid Motion Foundation Models with Large-Scale Human Videos — arXiv:2610.41877 — arXiv (en), https://arxiv.org/abs/2610.41877 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cb2b3a2c6524 (Metraipolis archive dc_cb2b3a2c6524, sha256 4fbaa196e62008cbf8701c07ea765402e36edf1fb25d0cf68622d7a68f6b6e80)
BibTeX
@misc{dccb2b3a2c6524,
  title = {Movens-Zero: Scaling Humanoid Motion Foundation Models with Large-Scale Human Videos — arXiv:2610.41877},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41877},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cb2b3a2c6524; sha256 4fbaa196e62008cbf8701c07ea765402e36edf1fb25d0cf68622d7a68f6b6e80},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41877. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Scaling humanoid motion foundation models is constrained by the scarcity of diverse, robot-executable data and by the difficulty of jointly learning from heterogeneous multimodal inputs while supporting real-time closed-loop control. We present Movens-5M, constructed from raw Internet videos through clip generation, quality filtering, semantic annotation, motion recovery, and physical refinement with humanoid retargeting and execution verification. It contains 15 million human-motion sequences and 5 million physically executable humanoid-motion sequences spanning over 10,000 hours, aligned with text, RGB video, speech, and music. Building on this data, we introduce Movens-Zero, a scalable understanding-generation-execution architecture that supports joint learning across modalities and real-time closed-loop execution. Task-mixed flow-matching training jointly optimizes a single action expert across seven generation tasks, enabling direct motion generation from text, speech, music, and video. Multimodal Real-Time Chunking connects overlapping chunks under continuously updated instructions and robot states for continuous closed-loop execution. We further introduce Movens-Intent, a four-track benchmark with 8,192 expert-verified condition-motion pairs across text, speech, music, and video. Experiments show consistent gains across all four modalities when scaling execution-verified data from 1% to 100% and action-expert capacity from 64M to 911M parameters, while demonstrating that data quality and cross-modal supervision improve generation and execution.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.41877 — 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 2026-11-03T05:33:32.339Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).