GeoCR: Learning a Generalist Cloud Removal Prior from Heterogeneous Observations
cs.CV
Submitted: 2026-09-26
Updated: 2026-09-26
Comments: Please visit our project page https://kaist-viclab.github.io/GeoCR_site/
Project page: https://kaist-viclab.github.io/GeoCR_site
License: http://creativecommons.org/licenses/by/4.0/
The gist: Cloud removal methods are typically specialized to individual datasets and input configurations, limiting reuse across sensors, spectral bands, and observation settings.
Terminology
Abstract
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.
Sources
- Demystifying MMD GANs
- A Unified Foundation Model for All-in-One Multi-Modal Remote Sensing Image Restoration and Fusion with Language Prompting
- LoRA: Low-Rank Adaptation of Large Language Models
- EO-VAE: Towards A Multi-sensor Tokenizer for Earth Observation Data
- A Remote Sensing Image Dataset for Cloud Removal
- Flow Matching for Generative Modeling
- DINOv3
- UniTS: Unified Spatio-Temporal Generative Model for Remote Sensing
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models