Metraipolis › archive › dc_f444dd5073bd

GeoCR: Learning a Generalist Cloud Removal Prior from Heterogeneous Observations — arXiv:2609.32510

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32510
License (as archived)
CC0-1.0
Archived
2026-10-07T23:50:57.211Z
SHA-256 of the archived text
863c3460bada0884356dd6871db1ff56b4c2b9b53fa4719e178ae0d036e687c1
Stable URL of this page
https://metraipolis.com/documento/dc_f444dd5073bd

How to cite

GeoCR: Learning a Generalist Cloud Removal Prior from Heterogeneous Observations — arXiv:2609.32510 — arXiv (en), https://arxiv.org/abs/2609.32510 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f444dd5073bd (Metraipolis archive dc_f444dd5073bd, sha256 863c3460bada0884356dd6871db1ff56b4c2b9b53fa4719e178ae0d036e687c1)
BibTeX
@misc{dcf444dd5073bd,
  title = {GeoCR: Learning a Generalist Cloud Removal Prior from Heterogeneous Observations — arXiv:2609.32510},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32510},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f444dd5073bd; sha256 863c3460bada0884356dd6871db1ff56b4c2b9b53fa4719e178ae0d036e687c1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32510. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Cloud removal methods are typically specialized to individual datasets and input configurations, limiting reuse across sensors, spectral bands, and observation settings. We introduce GeoCR, a generalist model that unifies RGB-only-based CR and multispectral-based CR from single- or multi-temporal cloudy observations, with optional SAR guidance, within a single network. To accommodate different spectral and sensing domains, compact input and output stems extend a pretrained RGB autoencoder while keeping its encoder and decoder trunks frozen. This shared latent interface enables a single flow transformer to jointly model clean RGB and non-RGB latents, conditioned on separate cloudy-observation streams and optional SAR tokens. Through joint pretraining on the training splits of ten datasets comprising 883,331 cloud-free target images, GeoCR learns a shared cloud removal prior across these heterogeneous configurations. The same pretrained checkpoint supports direct inference without dataset-specific fine-tuning and efficient adaptation through low-rank adaptation (LoRA). We evaluate GeoCR against general image restoration and cloud removal methods on test splits of the contributing datasets under full-band and RGB-only settings. GeoCR achieves the best FID and DISTS on full-band SEN12MS-CR and Sen2_MTC_New and RGB-only CUHK-CR2, outperforming existing models and demonstrating the effectiveness of a reusable generative model across diverse settings.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.32510 — 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 2026-10-07T23:29:27.054Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).