Metraipolis › archive › dc_a5ff9f2817bf

GeoFlip: Benchmarking Geometry-Sensitive Physics in Video Generation — arXiv:2610.42002

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.42002
License (as archived)
CC0-1.0
Archived
2026-11-02T05:04:19.163Z
SHA-256 of the archived text
be9657684664b13183a9c17b447611aad811ed7b8ef1428233148d45fa284293
Stable URL of this page
https://metraipolis.com/documento/dc_a5ff9f2817bf

How to cite

GeoFlip: Benchmarking Geometry-Sensitive Physics in Video Generation — arXiv:2610.42002 — arXiv (en), https://arxiv.org/abs/2610.42002 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a5ff9f2817bf (Metraipolis archive dc_a5ff9f2817bf, sha256 be9657684664b13183a9c17b447611aad811ed7b8ef1428233148d45fa284293)
BibTeX
@misc{dca5ff9f2817bf,
  title = {GeoFlip: Benchmarking Geometry-Sensitive Physics in Video Generation — arXiv:2610.42002},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.42002},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a5ff9f2817bf; sha256 be9657684664b13183a9c17b447611aad811ed7b8ef1428233148d45fa284293},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.42002. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Video generation models synthesize realistic motion, but can they respond correctly to changes in initial geometry? We introduce GeoFlip, a video generation benchmark centered on physical outcome flips induced by geometry. With shared physical settings and textual prompts held fixed, GeoFlip applies controlled interventions to support extent, object placement, and mechanism configuration, constructing paired scenes with contrasting reference events, such as remaining stable versus falling or tipping. These pairs test whether models translate geometric differences in the first frame into correct physical evolution. The benchmark comprises 482 scene pairs and 964 reference videos across six physical categories, with reference outcomes established through procedural scene construction, physics simulation, and numerical validation and verification of physical events. We evaluate 16 models for image to video generation along three dimensions: event correctness, structural integrity, and similarity to reference videos. We examine evaluation consistency with independent vision language judges. On the 442 conditions shared by all models, the highest event correctness averaged across tasks is 27.98%, revealing substantial limitations in physical generation conditioned on geometry. Model strengths vary across physical mechanisms, and the three evaluation dimensions capture distinct capabilities. Typical failures include ignoring geometric changes, generating incorrect object motion, bending rigid objects, and introducing additional components. Rank correlations between judges reach 0.906 for event correctness and 0.929 for structural integrity. By combining controlled geometric interventions, verifiable physical references, and multidimensional evaluation, GeoFlip assesses progress from visual realism toward physical consistency conditioned on geometry.

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/2610.42002 — 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 2026-11-02T04:49:13.281Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).