SAREO‑FM is a foundation model that separates semantic supervision from reconstruction supervision in synthetic aperture radar (SAR) and electro‑optical (EO) imagery. It uses modality tokens for masked reconstruction and learnable semantic queries guided by a pretrained vision foundation model to capture scene content. Trained on the million‑scale SAR‑1M corpus, the model shows strong unimodal transfer for SAR and EO alone and significant improvements when both modalities are combined.
arXiv:2510. 22665v4 Announce Type: replace-cross Abstract: Synthetic Aperture Radar (SAR) is a critical imaging modality due to its all-weather operational capability.
arXiv:2609.37496v1 Announce Type: new Abstract: Paired synthetic aperture radar (SAR) and electro-optical (EO) imagery is increasingly available across sensors, resolutions, and geographic regions. Y...
arXiv:2607.23238v3 Announce Type: replace Abstract: Masked image modeling has become a dominant paradigm for SAR pre-training, yet the design of the reconstruction target remains fundamentally unsett...
The paper introduces BMT, a unified hierarchical Vision Transformer that jointly performs SAR-to-optical image translation and semantic segmentation. It incorporates a LocalViTBlock, an enhanced output module, a ControlNet-style conditional injection, and a bounded Kendall uncertainty weighting scheme to balance the two tasks. Experiments on paired and unpaired datasets demonstrate competitive performance in both translation quality and segmentation accuracy.
arXiv:2609.00968v1 Announce Type: new Abstract: SAR-to-EO image translation aims to generate electro-optical (EO) imagery from synthetic aperture radar (SAR) observations. Existing latent diffusion a...
The paper proposes a cross‑modal framework that uses a frozen DINOv3 optical vision foundation model to create class‑level prototypes for Synthetic Aperture Radar (SAR) target recognition. By aligning SAR embeddings to these optical prototypes, the SAR model learns to classify SAR images without needing paired optical data. Experiments on the UNICORNv2 dataset show that this prototype alignment improves SAR classification accuracy compared to baseline methods and yields clearer class separation in the embedding space.
arXiv:2602. 19190v4 Announce Type: replace-cross Abstract: Research on the intelligent interpretation of all-weather, all-time Synthetic Aperture Radar (SAR) is crucial for advancing remote sensing applications.
C‑DiffSET is a SAR‑to‑EO image translation framework that uses a pretrained Latent Diffusion Model to adapt SAR imagery to the EO domain. The method exploits the pretrained VAE encoder’s ability to map SAR and EO images into a shared latent space, even when SAR inputs contain varying noise levels. A confidence‑guided diffusion loss further improves pixel‑wise fidelity by reducing artifacts such as appearing or disappearing objects, leading to state‑of‑the‑art results across multiple datasets.
Agricultural monitoring faces unique challenges, arising from the landscape's complex temporal, phenological, and climate dynamics, yet monitoring them is critical for ensuring food security. Synthetic Aperture Radar (SAR) satellites offer all-weather day-night imaging capability supporting key monitoring tasks including crop type mapping, yield prediction and phenological event detection.
ProSR is a new method for Synthetic Aperture Radar (SAR) super‑resolution that treats the task as a discrete token prediction problem in a quantized latent space. By mapping SAR signal features to discrete scattering primitives and using a self‑supervised backbone to extract label‑free semantic priors, ProSR preserves the impulsive, physically consistent scattering statistics of SAR images. The approach includes Semantic‑Aligned Detail Encoding and a Prototype‑Map‑Guided Attention mechanism, and it has been validated on a large‑scale 0.25 m resolution benchmark from the Umbra Open Dataset, achieving superior visual quality while maintaining essential scattering characteristics.
OptiSAR-Net++ introduces a new cross‑domain remote sensing visual grounding task (CD‑RSVG) and the first large‑scale benchmark dataset, OptSAR‑RSVG. The framework replaces Transformer decoding with a CLIP‑based contrastive approach, employing a patch‑level Low‑Rank Adaptation Mixture of Experts for efficient cross‑domain feature decoupling and a text‑guided dual‑gate fusion module for improved semantic‑visual alignment. Experiments show state‑of‑the‑art performance on OptSAR‑RSVG and DIOR‑RSVG, with notable gains in localization accuracy and computational efficiency.
The paper demonstrates a lightweight method to adapt general-purpose vision‑language models (VLMs) for multispectral and synthetic aperture radar (SAR) image understanding. By rendering each observation as five optical views and one SAR view, naming them in the prompt, and applying LoRA to the language network and selected visual transformer blocks, the authors enable VLMs to process band composites, spectral indices, and radar backscatter without retraining a new foundation model. On a balanced six‑class land‑cover benchmark from BigEarthNet‑v2, the adapted Qwen3‑VL achieves a micro F1 score of 0.8275, and the same protocol improves four other VLMs and transfers to flood verification and captioning tasks.