Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.23816 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914132159102976 |
|---|---|
| author | Fallah, Forouzan Li, Wenwen Hsu, Chia-Yu Lee, Hyunho Yang, Yezhou |
| author_facet | Fallah, Forouzan Li, Wenwen Hsu, Chia-Yu Lee, Hyunho Yang, Yezhou |
| contents | Super-resolution (SR) for remote sensing imagery often fails under out-of-distribution (OOD) conditions, such as rare geomorphic features captured by diverse sensors, producing visually plausible but physically inaccurate results. We present RareFlow, a physics-aware SR framework designed for OOD robustness. RareFlow's core is a dual-conditioning architecture. A Gated ControlNet preserves fine-grained geometric fidelity from the low-resolution input, while textual prompts provide semantic guidance for synthesizing complex features. To ensure physically sound outputs, we introduce a multifaceted loss function that enforces both spectral and radiometric consistency with sensor properties. Furthermore, the framework quantifies its own predictive uncertainty by employing a stochastic forward pass approach; the resulting output variance directly identifies unfamiliar inputs, mitigating feature hallucination. We validate RareFlow on a new, curated benchmark of multi-sensor satellite imagery. In blind evaluations, geophysical experts rated our model's outputs as approaching the fidelity of ground truth imagery, significantly outperforming state-of-the-art baselines. This qualitative superiority is corroborated by quantitative gains in perceptual metrics, including a nearly 40\% reduction in FID. RareFlow provides a robust framework for high-fidelity synthesis in data-scarce scientific domains and offers a new paradigm for controlled generation under severe domain shift. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_23816 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RareFlow: Physics-Aware Flow-Matching for Cross-Sensor Super-Resolution of Rare-Earth Features Fallah, Forouzan Li, Wenwen Hsu, Chia-Yu Lee, Hyunho Yang, Yezhou Computer Vision and Pattern Recognition Super-resolution (SR) for remote sensing imagery often fails under out-of-distribution (OOD) conditions, such as rare geomorphic features captured by diverse sensors, producing visually plausible but physically inaccurate results. We present RareFlow, a physics-aware SR framework designed for OOD robustness. RareFlow's core is a dual-conditioning architecture. A Gated ControlNet preserves fine-grained geometric fidelity from the low-resolution input, while textual prompts provide semantic guidance for synthesizing complex features. To ensure physically sound outputs, we introduce a multifaceted loss function that enforces both spectral and radiometric consistency with sensor properties. Furthermore, the framework quantifies its own predictive uncertainty by employing a stochastic forward pass approach; the resulting output variance directly identifies unfamiliar inputs, mitigating feature hallucination. We validate RareFlow on a new, curated benchmark of multi-sensor satellite imagery. In blind evaluations, geophysical experts rated our model's outputs as approaching the fidelity of ground truth imagery, significantly outperforming state-of-the-art baselines. This qualitative superiority is corroborated by quantitative gains in perceptual metrics, including a nearly 40\% reduction in FID. RareFlow provides a robust framework for high-fidelity synthesis in data-scarce scientific domains and offers a new paradigm for controlled generation under severe domain shift. |
| title | RareFlow: Physics-Aware Flow-Matching for Cross-Sensor Super-Resolution of Rare-Earth Features |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.23816 |