Metraipolis › archive › dc_31d8341e6366

HyperbolicDiffusion: Sharp & Scalable Tiled Generation on the Hyperbolic Plane — arXiv:2608.03422

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.03422
License (as archived)
CC0-1.0
Archived
2026-10-31T12:58:26.336Z
SHA-256 of the archived text
aaf443fb8beb00279903ae67031e4dc30f6824593e5b4c391aa92e1ea8702c97
Stable URL of this page
https://metraipolis.com/documento/dc_31d8341e6366

How to cite

HyperbolicDiffusion: Sharp & Scalable Tiled Generation on the Hyperbolic Plane — arXiv:2608.03422 — arXiv (en), https://arxiv.org/abs/2608.03422 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_31d8341e6366 (Metraipolis archive dc_31d8341e6366, sha256 aaf443fb8beb00279903ae67031e4dc30f6824593e5b4c391aa92e1ea8702c97)
BibTeX
@misc{dc31d8341e6366,
  title = {HyperbolicDiffusion: Sharp & Scalable Tiled Generation on the Hyperbolic Plane — arXiv:2608.03422},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.03422},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_31d8341e6366; sha256 aaf443fb8beb00279903ae67031e4dc30f6824593e5b4c391aa92e1ea8702c97},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03422. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Planar tiled diffusion denoises overlapping windows of one rectangular canvas. The hyperbolic plane has no such canvas, and its area grows exponentially with radius. We introduce HyperbolicDiffusion, a training-free method for generating finite visual fields directly on the hyperbolic plane H2. Our Hyperbolic Blooming Cover reduces window placement to a compact dynamic program that runs in seconds while providing strong theoretical guarantees. Permanent surface IDs form a shared latent canvas: a standard diffusion model denoises local windows, whose predictions are fused back onto H2. Because curvature causes residual disagreement and blur at multi-window junctions, a geometry-derived second stage re-noises and repairs precisely those regions. The resulting fields are sharp, reprojectable, and consistent across viewpoints, providing a prompt-driven generative counterpart to Escher's Circle Limit series.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-05 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.03422 — 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 2026-10-31T12:45:33.249Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).