Metraipolis › archive › dc_cd55a4e432e7
Atomizer-IO: Beyond Pixels, Patches and Grids — arXiv:2609.40320
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.40320
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T14:46:40.216Z
- SHA-256 of the archived text
2a8eb4908b2e514f49427693225e4dbf1ba14c894d8b613997e3a96efd68b15b- Stable URL of this page
https://metraipolis.com/documento/dc_cd55a4e432e7
How to cite
Atomizer-IO: Beyond Pixels, Patches and Grids — arXiv:2609.40320 — arXiv (en), https://arxiv.org/abs/2609.40320 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cd55a4e432e7 (Metraipolis archive dc_cd55a4e432e7, sha256 2a8eb4908b2e514f49427693225e4dbf1ba14c894d8b613997e3a96efd68b15b)
BibTeX
@misc{dccd55a4e432e7,
title = {Atomizer-IO: Beyond Pixels, Patches and Grids — arXiv:2609.40320},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.40320},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cd55a4e432e7; sha256 2a8eb4908b2e514f49427693225e4dbf1ba14c894d8b613997e3a96efd68b15b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.40320. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Most vision architectures assume that observations lie on a regular grid, an effective abstraction for natural images but a restrictive one for sensing data whose channels, temporal sampling, spatial resolution, and geometry can vary. Generic set-based architectures remove the grid, but also remove useful spatial inductive biases. We introduce Atomizer-IO, an architecture that places observations first and derives structure from their physical relationships. Building on top of an atomic representation of the data, each observation is described by its measurement and acquisition metadata, while local cross-attention maps observations to anchor points that can be arbitrarily placed. We evaluate this design by progressively relaxing the grid assumption, from varying input raster configurations and incomplete channel sets to flexible output density and, ultimately, inputs without a raster grid. Atomizer-IO is competitive with flexible EO-specific architectures on most tasks, while offering post-training control over inference cost and competitive compute--performance trade-offs. The same formulation extends without architectural redesign to unordered 3D point clouds, showing that the atomic interface generalizes beyond regular raster inputs. These results suggest that pixels, patches, and grids do not need to define the interface of a sensing architecture. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.40320 — 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 2026-10-07T14:19:00.861Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).