Search papers, labs, and topics across Lattice.
This paper introduces Geo-Anchored Cloud Removal (GACR), a novel framework for cloud removal in optical remote sensing that prioritizes both visual realism and downstream interpretability. By employing Observation-Anchored Residual Flow (OAR-Flow), GACR reformulates the cloud removal process to anchor the generative trajectory to the actual cloudy observations, resulting in faster and more stable reconstructions. The integration of Geo-Contextual Prior Alignment (GCPA) ensures that semantic structures are preserved, leading to significant improvements in reconstruction quality and downstream task performance across multiple datasets.
Cloud removal techniques can now enhance downstream task accuracy by preserving spatial-semantic integrity, not just visual fidelity.
Cloud removal (CR) is essential for optical remote sensing, serving as a prerequisite for reliable downstream interpretation, such as semantic segmentation and change detection. However, existing CR approaches often prioritize visual realism while overlooking their impact on subsequent analytical tasks, leading to semantic drift and degraded downstream performance. To address this issue, we propose Geo-Anchored Cloud Removal (GACR), a unified framework that jointly ensures faithful reconstruction and robust interpretability. At its core, GACR incorporates Observation-Anchored Residual Flow (OAR-Flow), which reformulates CR as a physically grounded residual inversion process. By anchoring the generative trajectory to the cloudy observation rather than pure noise, OAR-Flow enables fast, stable, and faithful reconstruction. To further preserve semantic structures critical for downstream interpretation, GACR integrates Geo-Contextual Prior Alignment (GCPA) to constrain the reconstruction within a semantic manifold induced by a Vision Foundation Model (VFM). Consequently, GACR strictly maintains the spatial-semantic integrity of complex landscapes. Extensive experiments across six CR datasets and twelve downstream tasks demonstrate that GACR produces superior reconstruction quality while consistently improving downstream task accuracy. The code is available at https://github.com/wzy6055/GACR.
