Metraipolis › archive › dc_acc88015fb73

Parasitic Co-Denoising: Unlocking 3D Human Motion Generation in a Frozen Video Diffusion Model — arXiv:2610.03047

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.03047
License (as archived): CC0-1.0
Archived: 2026-10-06T09:07:38.519Z
SHA-256 of the archived text: d4b94d28f41ab90ecb2136054563d0481bf39f6573f054afb79e3c953d784417
Stable URL of this page: https://metraipolis.com/documento/dc_acc88015fb73

How to cite

Parasitic Co-Denoising: Unlocking 3D Human Motion Generation in a Frozen Video Diffusion Model — arXiv:2610.03047 — arXiv (en), https://arxiv.org/abs/2610.03047 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_acc88015fb73 (Metraipolis archive dc_acc88015fb73, sha256 d4b94d28f41ab90ecb2136054563d0481bf39f6573f054afb79e3c953d784417)

BibTeX

@misc{dcacc88015fb73,
  title = {Parasitic Co-Denoising: Unlocking 3D Human Motion Generation in a Frozen Video Diffusion Model — arXiv:2610.03047},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.03047},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_acc88015fb73; sha256 d4b94d28f41ab90ecb2136054563d0481bf39f6573f054afb79e3c953d784417},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.03047. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Despite never being supervised on explicit 3D motion, large-scale text-to-video diffusion models synthesize realistic human motion in their generated videos. We ask whether this implicit knowledge can be turned into explicit 3D motion generation, without training a separate motion model. Probing a frozen Wan2.1 reveals that a recoverable motion signal is present in its intermediate states across the entire denoising schedule, not confined to the clean output. Motivated by this, we introduce parasitic co-denoising, a paradigm in which motion is decoded from the host model along its denoising schedule rather than produced by an independent generator. We instantiate it as the Parasitic Motion Decoder (PMD), an efficient flow-matching decoder that shares the host's noise schedule and reads its intermediate features through a $\sigma$-adaptive multi-layer fusion, leaving the host unmodified. Drawing its coverage from the host rather than from motion data, PMD leads dedicated motion generators on text-motion alignment at a small fraction of their trainable parameters, while producing paired video and motion in a single pass that motion-only baselines cannot match.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-05 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.03047 — 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 2026-10-06T08:21:44.198Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).