Problem
Research questions and friction points this paper is trying to address.
Remote sensing image restoration
Multiple degradations
Heterogeneous data
Multi-task learning
Unified framework
Task-Guided Prompting for Unified Remote Sensing Image Restoration
🤖 AI Summary
Existing remote sensing image restoration methods typically address only a single degradation type or homogeneous data, struggling to handle the complex challenges of coupled degradations and cross-modal variations in real-world scenarios. This work proposes TGPNet, a unified restoration framework that introduces a novel task-guided prompting mechanism: learnable task embeddings generate degradation-aware prompts, which, combined with hierarchical feature modulation, enable adaptive processing of five distinct tasks—denoising, cloud removal, shadow elimination, deblurring, and SAR despeckling—under shared network weights. Evaluated on a newly constructed multimodal unified remote sensing restoration benchmark, TGPNet not only excels in composite degradation settings but also surpasses multiple specialized models, demonstrating its strong generalization capability and state-of-the-art performance.
📝 Abstract
Remote sensing image restoration (RSIR) is essential for recovering high-fidelity imagery from degraded observations, enabling accurate downstream analysis. However, most existing methods focus on single degradation types within homogeneous data, restricting their practicality in real-world scenarios where multiple degradations often across diverse spectral bands or sensor modalities, creating a significant operational bottleneck. To address this fundamental gap, we propose TGPNet, a unified framework capable of handling denoising, cloud removal, shadow removal, deblurring, and SAR despeckling within a single, unified architecture. The core of our framework is a novel Task-Guided Prompting (TGP) strategy. TGP leverages learnable, task-specific embeddings to generate degradation-aware cues, which then hierarchically modulate features throughout the decoder. This task-adaptive mechanism allows the network to precisely tailor its restoration process for distinct degradation patterns while maintaining a single set of shared weights. To validate our framework, we construct a unified RSIR benchmark covering RGB, multispectral, SAR, and thermal infrared modalities for five aforementioned restoration tasks. Experimental results demonstrate that TGPNet achieves state-of-the-art performance on both unified multi-task scenarios and unseen composite degradations, surpassing even specialized models in individual domains such as cloud removal. By successfully unifying heterogeneous degradation removal within a single adaptive framework, this work presents a significant advancement for multi-task RSIR, offering a practical and scalable solution for operational pipelines. The code and benchmark will be released at https://github.com/huangwenwenlili/TGPNet.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Task-Guided Prompting
Unified Restoration
Remote Sensing Image Restoration
Multi-Modal
Degradation-Aware Modulation
🔎 Similar Papers
No similar papers found.
