Metraipolis › archive › dc_465f2ea75d2d

TourPhysics: Bringing Physics to World Models for Exploration and Manipulation from a Single Image — arXiv:2609.04911

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.04911
License (as archived): CC0-1.0
Archived: 2026-10-16T23:41:14.079Z
SHA-256 of the archived text: 41ca8e99b8fc365098e5233ca9cd1fa0b05747e80c7cab4f83b87c4ca647be76
Stable URL of this page: https://metraipolis.com/documento/dc_465f2ea75d2d

How to cite

TourPhysics: Bringing Physics to World Models for Exploration and Manipulation from a Single Image — arXiv:2609.04911 — arXiv (en), https://arxiv.org/abs/2609.04911 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_465f2ea75d2d (Metraipolis archive dc_465f2ea75d2d, sha256 41ca8e99b8fc365098e5233ca9cd1fa0b05747e80c7cab4f83b87c4ca647be76)

BibTeX

@misc{dc465f2ea75d2d,
  title = {TourPhysics: Bringing Physics to World Models for Exploration and Manipulation from a Single Image — arXiv:2609.04911},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.04911},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_465f2ea75d2d; sha256 41ca8e99b8fc365098e5233ca9cd1fa0b05747e80c7cab4f83b87c4ca647be76},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04911. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Interactive visual world models must distinguish observation from physical intervention. Camera motion reveals new surfaces, whereas intervention changes object motion, contact, and deformation. Current video world models are largely driven by appearance priors and often lose physical or spatial consistency over long horizons. We present TourPhysics, an online framework initialized from a single image and a declarative physical configuration. TourPhysics extends PhysOmni, our ACM Multimedia 2026 work, from finite physics-grounded video synthesis to persistent exploration and manipulation. TourPhysics combines deterministic simulation with video generation while assigning separate roles to simulator state, geometric evidence, generator controls, and appearance memory. For each action, the simulator computes a finite physical and camera trajectory before the corresponding observation is generated. Accepted observations publish the terminal state and update the appearance memory and subsequent generator controls, while the committed state and simulator geometry remain fixed throughout synthesis and retry. We further separate the simulator geometry used for projection and visibility from the relative depth used to condition the generator. A reference-anchored memory retrieves accepted static appearance through geometric cross-view correspondence and incorporates it through a bounded residual that reverts to the native path when no valid correspondence exists. On simulator-defined camera tours and object manipulations, TourPhysics follows prescribed camera and object trajectories more closely than the evaluated baselines, preserves the input scene, and reduces appearance drift during long-horizon revisits.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-07 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.04911 — 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 2026-10-16T23:16:11.420Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).