Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

UniRoute: Unified Routing Mixture-of-Experts for Modality-Adaptive Remote Sensing Change Detection

Qingling Shu, Sibao Chen, Wei Lu, Zhihui You, Chengzhuang Liu

TL;DR

UniRoute tackles modality-adaptive remote sensing change detection by unifying feature extraction and fusion through conditional routing. It introduces AR2-MoE to disentangle local detail versus global context and MDR-MoE to tailor fusion primitives to each modality, complemented by CASD to stabilize training on data-scarce heterogeneous data. The approach achieves strong cross-dataset performance with a single, efficient model, outperforming unified baselines and approaching specialist ensembles in accuracy while using substantially fewer parameters and FLOPs. The work advances practical deployment of change detection across diverse sensing modalities by providing a scalable, robust framework with principled regularization.

Abstract

Current remote sensing change detection (CD) methods mainly rely on specialized models, which limits the scalability toward modality-adaptive Earth observation. For homogeneous CD, precise boundary delineation relies on fine-grained spatial cues and local pixel interactions, whereas heterogeneous CD instead requires broader contextual information to suppress speckle noise and geometric distortions. Moreover, difference operator (e.g., subtraction) works well for aligned homogeneous images but introduces artifacts in cross-modal or geometrically misaligned scenarios. Across different modality settings, specialized models based on static backbones or fixed difference operations often prove insufficient. To address this challenge, we propose UniRoute, a unified framework for modality-adaptive learning by reformulating feature extraction and fusion as conditional routing problems. We introduce an Adaptive Receptive Field Routing MoE (AR2-MoE) module to disentangle local spatial details from global semantic context, and a Modality-Aware Difference Routing MoE (MDR-MoE) module to adaptively select the most suitable fusion primitive at each pixel. In addition, we propose a Consistency-Aware Self-Distillation (CASD) strategy that stabilizes unified training under data-scarce heterogeneous settings by enforcing multi-level consistency. Extensive experiments on five public datasets demonstrate that UniRoute achieves strong overall performance, with a favorable accuracy-efficiency trade-off under a unified deployment setting.

UniRoute: Unified Routing Mixture-of-Experts for Modality-Adaptive Remote Sensing Change Detection

TL;DR

UniRoute tackles modality-adaptive remote sensing change detection by unifying feature extraction and fusion through conditional routing. It introduces AR2-MoE to disentangle local detail versus global context and MDR-MoE to tailor fusion primitives to each modality, complemented by CASD to stabilize training on data-scarce heterogeneous data. The approach achieves strong cross-dataset performance with a single, efficient model, outperforming unified baselines and approaching specialist ensembles in accuracy while using substantially fewer parameters and FLOPs. The work advances practical deployment of change detection across diverse sensing modalities by providing a scalable, robust framework with principled regularization.

Abstract

Current remote sensing change detection (CD) methods mainly rely on specialized models, which limits the scalability toward modality-adaptive Earth observation. For homogeneous CD, precise boundary delineation relies on fine-grained spatial cues and local pixel interactions, whereas heterogeneous CD instead requires broader contextual information to suppress speckle noise and geometric distortions. Moreover, difference operator (e.g., subtraction) works well for aligned homogeneous images but introduces artifacts in cross-modal or geometrically misaligned scenarios. Across different modality settings, specialized models based on static backbones or fixed difference operations often prove insufficient. To address this challenge, we propose UniRoute, a unified framework for modality-adaptive learning by reformulating feature extraction and fusion as conditional routing problems. We introduce an Adaptive Receptive Field Routing MoE (AR2-MoE) module to disentangle local spatial details from global semantic context, and a Modality-Aware Difference Routing MoE (MDR-MoE) module to adaptively select the most suitable fusion primitive at each pixel. In addition, we propose a Consistency-Aware Self-Distillation (CASD) strategy that stabilizes unified training under data-scarce heterogeneous settings by enforcing multi-level consistency. Extensive experiments on five public datasets demonstrate that UniRoute achieves strong overall performance, with a favorable accuracy-efficiency trade-off under a unified deployment setting.
Paper Structure (16 sections, 14 equations, 4 figures, 4 tables)

This paper contains 16 sections, 14 equations, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Remote Sensing Change Detection
  4. Mixture-of-Experts in Computer Vision
  5. Unified Modeling & Training Strategies
  6. Methodology
  7. Overview
  8. Adaptive Receptive Field Routing MoE
  9. Modality-Aware Difference Routing MoE
  10. Consistency-Aware Self-Distillation (CASD)
  11. Experiments and Analysis
  12. Datasets and Implementation Details
  13. Comparison with State-of-the-Art Methods
  14. Ablation Study
  15. Generalization and Data Efficiency
  16. ...and 1 more sections

Figures (4)

  • Figure 1: (a) Isolated specialized methods for different modality pairs. (b) Static unified baselines suffering from fixed receptive fields and indiscriminate fusion. (c) Our UniRoute framework with modality-adaptive routing and consistency-aware training.
  • Figure 2: Overview of UniRoute. It features an $\text{AR}^2$-MoE for adaptive receptive field selection in the encoder and an MDR-MoE for optimal fusion primitive routing in the decoder, optimized via the CASD strategy.
  • Figure 3: Qualitative comparisons on five datasets. We show predicted change maps of representative methods and our UniRoute on both homogeneous and heterogeneous settings.
  • Figure 4: t-SNE visualization of feature distributions.