Compositional-Degradation UAV Image Restoration: Conditional Decoupled MoE Network and A Benchmark

2026-04-10

AI Summary

This work addresses the challenge of restoring drone-captured images degraded by complex, real-world factors such as rain, fog, and noise, where existing unified restoration methods suffer from mutual interference due to implicit degradation modeling. To overcome this, we propose DAME-Net, which explicitly perceives individual degradation factors and decouples their reconstruction processes for efficient restoration. Our key innovations include a novel factor-level degradation-aware module and a conditionally decoupled Mixture-of-Experts (MoE) architecture that replaces implicit conditioning with explicit, interpretable degradation descriptors. We also introduce MDUR, the first large-scale benchmark for composite-degradation drone image restoration, along with a multi-label soft-alignment perception mechanism and a spatial-frequency hybrid expert routing strategy. Experiments demonstrate that DAME-Net significantly outperforms state-of-the-art methods on MDUR, particularly under unseen and higher-order composite degradations, while also enhancing downstream object detection performance.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Mixture of Experts (MoE)
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

UAV images are critical for applications such as large-area mapping, infrastructure inspection, and emergency response. However, in real-world flight environments, a single image is often affected by multiple degradation factors, including rain, haze, and noise, undermining downstream task performance. Current unified restoration approaches typically rely on implicit degradation representations that entangle multiple factors into a single condition, causing mutual interference among heterogeneous corrections. To this end, we propose DAME-Net, a Degradation-Aware Mixture-of-Experts Network that decouples explicit degradation perception from degradation-conditioned reconstruction for compositional UAV image restoration. Specifically, we design a Factor-wise Degradation Perception module(FDPM) to provide explicit per-factor degradation cues for the restoration stage through multi-label prediction with label-similarity-guided soft alignment, replacing implicit entangled conditions with interpretable and generalizable degradation descriptions. Moreover, we develop a Conditioned Decoupled MoE module(CDMM) that leverages these cues for stage-wise conditioning, spatial-frequency hybrid processing, and mask-constrained decoupled expert routing, enabling selective factor-specific correction while suppressing irrelevant interference. In addition, we construct the Multi-Degradation UAV Restoration benchmark (MDUR), the first large-scale UAV benchmark for compositional UAV image restoration, with 43 degradation configurations from single degradations to four-factor composites and standardized seen/unseen splits.Extensive experiments on MDUR demonstrate consistent improvements over representative unified restoration methods, with greater gains on unseen and higher-order composite degradations. Downstream experiments further validate benefits for UAV object detection.

Problem

• UAV image restoration
• compositional degradation
• multi-degradation
• image restoration
• degradation decoupling

Innovation

• Degradation Decoupling
• Mixture-of-Experts
• UAV Image Restoration
• Composite Degradation
• Explicit Degradation Perception