GeoCR：从异构观测中学习通用云去除先验
GeoCR: Learning a Generalist Cloud Removal Prior from Heterogeneous Observations
中文总结 AI 辅助
GeoCR通过联合预训练十个数据集，学习跨RGB和多光谱的通用云去除先验，支持直接推理和LoRA适配，并在多个基准上取得最优FID和DISTS。
中文摘要 AI 辅助
云去除方法通常针对单个数据集和输入配置进行专门设计，限制了其在传感器、光谱波段和观测设置之间的复用。我们提出了GeoCR，一种通用模型，它在单个网络中统一了基于RGB-only的云去除和基于多光谱的云去除，可处理单时相或多时相的有云观测，并可选地使用SAR引导。为了适应不同的光谱和感知域，紧凑的输入和输出分支扩展了预训练的RGB自编码器，同时保持其编码器和解码器主干冻结。这种共享的潜在接口使单个流变换器能够联合建模干净的RGB和非RGB潜在表示，并以分离的有云观测流和可选的SAR令牌为条件。通过在十个数据集（包含883,331张无云目标图像）的训练划分上进行联合预训练，GeoCR学习了跨这些异构配置的共享云去除先验。相同的预训练检查点支持无需数据集特定微调的直接推理，以及通过低秩适配（LoRA）进行高效适配。我们在贡献数据集的测试划分上，在全波段和RGB-only设置下，将GeoCR与通用图像恢复和云去除方法进行了评估。GeoCR在全波段的SEN12MS-CR和Sen2_MTC_New以及RGB-only的CUHK-CR2上取得了最佳的FID和DISTS，优于现有模型，证明了可复用生成模型在不同设置下的有效性。
英文摘要
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.
